HRD-81-37 Implementing GAO's Recommendations on … · Implementing GAO’s Recommendations On The Social Security Administration’s Programs Could Save Billions ... HRD-81-37 DECEMBER

BY THE COMPTRtlLLER GENE!RiL

lM-+- *A /., 16 hida

‘k&port To The Congress OF THE UNITED STATES

Implementing GAO’s Recommendations On The Social Security Administration’s Programs Could Save Billions

This report summarizes actions taken on GAO’s recommendations in 27 previous reports on the Social Security Administration’s income security programs.

GAO believes the savings which would result from fully implementing its recommendations could help reduce the Federal deficit, reduce inflationary pressure, and improve the equity and integrity of income security programs. The recommendations that would produce the most substantial savings would affect the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds and would require changes to the Social Security Act.

Some of these recommendations have previously been considered by the Congress. How- ever, the Congress should reconsider GAO’s recommendations in view of the continuing emphasis on reducing the budget deficit, controlling inflation, and resolving the serious financial problems of the Old-Age and Sur- vivors Insurance Trust Fund.

II I 114063

HRD-81-37 DECEMBER 31,198O

For sale by:

Superintendent of Documents U.S. Government Printing Office Washington, D.C. 20402

Telephone (202) 783-3238

Members of Congress; heads of Federal, State, and local government agencies; members of the press; and libraries can obtain GAO documents from:

U.S. General Accounting Office Document Handling and Information

Services Facility P.O. Box 6015 Gaithersburg, Md. 20760

Telephone (202) 275-6241

COMPTROLLER OENERAL OF THE UNITED STATES WABHINQTON, D.C. 2054B

B-201375

To the President of the Senate and the Speaker of the House of Representatives

During the last several years, we have issued numerous reports recommending changes in income security programs managed by the Social Security Administration (SSA). Actions taken by the Congress and the Executive Branch on the matters discussed in our reports will save about $2 billion during fiscal years 1982-85. There would be additional savings of about $1.3 billion in fiscal year 1982 and about $4.5 billion in fiscal years 1983-85 if our recommendations were fully implemented. There would also be substantial savings in later years.

Legislative changes are needed to implement the recommendations that would result in the greatest savings.. Some of these recommendations have been considered by the Con- gress. However, we believe further consideration is warranted as part of congressional efforts to reduce the budget deficit, control inflation, and resolve the serious problems of the Old-Age and Survivors Insurance Trust Fund.

These matters are discussed below.

BACKGROUND

Income security programs account for over one-third of the Federal budget. SSA, within the Department of Health and Human Services (HHs), l/ administers some of the largest income security programs, including Old-Age and Survivors Insurance, Disability Insurance, Supplemental Security Income (SSI) I and Aid to Families with Dependent Children (AFDC). We have issued many reports on these programs containing

l/On May 4, 1980, the part of the Department of Health, Education, and Welfare (HEW) responsible for most of the activities discussed in this report became HHS.

B-201375 1

recommendations which, if implemented, could save billions of dollars. We made this review to determine what actions have been taken and what still needs to be done.

We obtained information from HHS on the actions taken in response to our recommendations and evaluated the adequacy of these actions. The actual or potential savings data were obtained from HHS. In some cases HHS projected savings through fiscal year 1986, but in other instances, data were readily available only for shorter time periods. In a few instances we made projections based on available data. We did not verify the savings data provided.

We also reviewed recent legislation and the status of legislative proposals to determine progress in implementing our legislative recommendations.

Appendix I summarizes our reports on the Old-Age and Survivors and Disability Insurance programs, appendix II summarizes our reports on the SSI program, and appendix III summarizes our reports on the AFDC and Emergency Assistance programs. Minor modifications have been made to some recommendations to recognize actions taken by the Congress or HHS to partially implement our recommendations.

We did not obtain written comments from HHS because this report principally reiterates those unresolved recommendations made in our prior reports and HHS has previously responded to those recommendations. However, drafts of the appendixes have been reviewed by SSA officials and revised, where appropriate, to reflect their comments.

WHY OUR RECOMMENDATIONS ARE IMPORTANT

Many believe that deficit Federal spending is a cause of or a major contributing factor to inflation. In his fiscal year 1981 budget message, the President stated that equitable budget restraint was essential to control inflation. The President's message cited several efforts needed to combat inflation, such as reducing dependence on foreign oil and enhancing economic productivity. However, he said that none of these efforts can succeed unless Federal spending is controlled.

2

B-201375

Also, although the largest income security programs are funded by payroll taxes rather than general revenues, serious financial difficulties are anticipated for the largest trust fund. The 1980 annual report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Disability Insur- ance Trust Funds states that, under present conditions, the assets of the Old-Age and Survivors Insurance Trust Fund are expected to be insufficient to pay benefits by late 1981 or early 1982.

In January 1980, the administration requested temporary authority for interfund borrowing, which would make the Old-Age and Survivors Insurance Trust Fund, the Disability Insurance Trust Fund, and the Hospital Insurance Trust Fund available to pay benefits under any of the three programs. Depending on the future course of the economy, interfund borrowing may be adequate to finance all three programs during the 1980s. Under adverse economic conditions, interfund borrowing would postpone but not eliminate the need for more income. Recent increases in social security taxes met with worker resistance and public debate as to whether to rescind the increases.

Finally, publicity about payments to ineligibles, overpayments, duplicate payments, and payments to people who are not needy have raised public concern.

Implementing our recommendations would help reduce inflation, enhance the financial condition of the trust funds, and improve the equity and integrity of income security programs.

ACTIONS TAKEN WILL SAVE ABOUT $2 BILLION IN FISCAL YEARS 1982-85

Action has been taken on the matters discussed in nine of our reports which will save about $2 billion during fiscal years 1982-85. For example, in March 1979, SSA instituted a procedure to verify that student beneficiaries were attending school full time as required to be eligible for student benefits. We estimate that this procedure will save about $36 million a year. In connection with our report on liberal deposit requirements of States' Social Security contributions, the Social Security Disability Amendments of 1980 (Public Law 96-265) was approved on June 19, 1980, which require States to deposit their contributions monthly instead of quarterly.

3

: 1.

,: .*_ ,I,’

‘fi.‘ .’

1, , I , .

; . ;

B-201375

This change is expected to result in additional interest of about $1.3 billion being earned during fiscal years 1982-85 and made available to the Social Security trust funds.

Also, in response to our recommendations, SSA has begun to (1) obtain pension and compensation data from the Department of Labor, Railroad Retirement Board, Veterans Administration, and Office of Personnel Management and (2) compare these data with those reported by SSI applicants and recipients to assure that payments are correct. We estimate that this action will save the Federal Government about $244 million during fiscal years 1982-85. The table on the following page summarizes these and other achieved savings and refers to pages in the appendixes which describe in greater detail the basis for the anticipated savings.

It was not practicable to precisely quantify by fiscal years the savings achieved through implementing many of our recommendations. However, it is apparent that implementing these recommendations has resulted in large savings, in addition to the quantified savings discussed above. For example, actions have been taken to improve the collection of SSI overpayments, but i,t was not possible to quantify the savings attributable to these actions. (See p. 39.)

FURTHER ACTIONS CAN SAVE BILLIONS MORE

Fully implementing the recommendations in eight of our reports would save $1.3 billion in fiscal year 1982 and about $4.5 billion in fiscal years 1983-85. There would be similar savings in later years. In addition, there would be substantial, although unquantifiable, savings from implementing many other recommendations.

Following are examples of programs where our recommendations, if implemented, would result in substantial savings:

--Terminating Social Security benefits for postsecondary students effective fall 1981 would result in net savings of about $1.1 billion during fiscal year 1982 and recurring savings of similar or larger amounts in later years. As an alternative, if student benefits were gradually phased out beginning with fiscal

4

B-201375 Appendix

page Fiscal years -___ Action takcsn reference 1982 1983 1984 1985 Total ----._I

(millions)

Student Social Security benefits: verify- ing full-time status

Student basic grants: interface with Social Security benefits

Expedited deposits of States' Social Security contributions

Improved collection of Social Security overpayments

Reduced SSI benefits: interface with other Federal agencies

Reduced SSI windfall benefits

Reduced SSX payments to newly arrived aliens

Reduced SSI disability payments to ineligibles

Reduced AFDC payments to ineligibles

Discontinuance of advance AFDC payments

Total

P* 1 $ 36.0 $ 36.0 $ 36.0 $ 36.0 $ 144.0

P* 1 5.8 5.8 5.8 5.8 23.2

PO 6 235.0 286.0 337.0 412.0 1,270.O

p. 17 8.7 8.7 8.7 8.7 34.0

p. 23 61.0 61.0 61.0 61.0 244.0

p. 25 24.0 30.0 33.0 35.0 122.0

p. 30 12.0 19.0 24.0 28.0 83.0

p. 34 2.4 2.2

7.1

(a) (6) a/4.6

p. 51 7.1 7.1 7.1 28.4

p. 56 0.8 0.8 0.8 0.8 3.2

$392.8 $456.6 $513.4 $594.4 $1,957.2

s/SSA did not have sufficient data to enable us to project savings for fiscal years 1984 and 1985.

5

B-201375

year 1982, the estimated net savings would be $74 million in fiscal year 1982 and additional savings of about $2.4 billion during fiscal years 1983-85.

--Eliminating the minimum Social Security benefit would result in estimated net savings of $35 million in fiscal year 1982 and additional savings of $240 million during fiscal years 1983-85.

--Requiring States to make more frequent Social Security deposits would earn additional interest of about $49 million in fiscal year 1982 and about $290 million during fiscal years 1983-85.

--Calculating Social Security benefits to the nearest penny --or to the nearest dime, as HHS has proposed-- would save at least $8 million in fiscal year 1982 and at least $181 million during fiscal years 1983-85.

The table on the following page shows the quantifiable savings that would result from implementing our recommendations.

It was not practicable to determine the savings on a fiscal year basis that would result from implementing many of our recommendations. However, we believe that taking action on such recommendations can result in large savings, in addition to the quantifiable savings discussed above.

CONCLUSIONS AND RECOMMENDATIONS

We believe the savings which would result from fully implementing our recommendations could help reduce the Federal deficit, reduce inflationary pressure, and improve the equity and integrity of income security programs. The recommendations that would produce the most substantial savings would affect the Federal Old-Age and Survivors Insurance and Disability Insurance Trust Funds and would require changes to the Social Security Act.

Some of these recommendations have previously been considered by the Congress. However, we believe the Congress should reconsider our recommendations in view of the continuing emphasis on reducing the budget deficit, controlling

6

Recommendation

Terminate student benefits

Reduce excessive student basic grants

Further expedite deposits of States' Social Security contributions

Rounding benefits to nearest penny

4 Eliminate minimum benefits

Use State data to compute benefits

Reduce SSI payments to newly arrived aliens

Prepayment review of retroactive SSI payments

Compute SSI benefits on a retrospective basis

Total

Appendix page

reference

P* 1

P* 1

P= 6

PO 8

p. 10

p. 20

p. 30

p. 42

p. 45

Fiscal years --~- .-------

- - 1982 1983 1984 1985 Total' Y --- ----_ -td

d -------------------(millions)---------------------- E

$1,120.0 $1,120.0

2.4 2.4 2.4 2-4 9.6

49.0

8.0

35.0

1.6

32.0 24.0 15.0 15.0 86.0

$1,120.0 $1,259.0 $4,619-O 2t

81.0 95.0 114.0

37.0 62.0 82.0

60.0 80.0 100 .o

1.6 1.6 1.6

2.0 2.0 2.0 2.0 8.0

60.0 60.0

$1,310.0 $1,388.0

339.0

189.0

275.0

6.4

60.0 60.0 240.0 ---

$1,438.0 $1,636.0 $5,772.0 --

B-201375

inflation, and resolving the serious financial problems of the Old-Age and Survivors Insurance Trust Fund.

Each of our recommendations to the Congress is stated below and referenced to the pertinent pages of the appendixes. For brevity, we are not restating our recommendations to the Secretaries of HHS, Education, or State. Those recommendations are in the appendixes.

RECOMMENDATIONS TO THE CONGRESS --

We recommend that the Congress amend the Social Security Act to:

--Discontinue student benefits for postsecondary students and take steps to assure that the Department of Educa- tion will have sufficient financial resources to meet any increased demand for aid arising from discontinuance of these benefits. (See pm 1.)

-Require States to deposit Social Security taxes semimonthly or biweekly. We also recommend that the Con- gress consider requiring States to deposit Social Security taxes using the same schedule that States now use to deposit withheld income taxes. Such a requirement would enable the trust funds to earn additional interest income over the $339 million which could be earned by requiring remittances semimonthly or biweekly. (See p. 6.)

--Compute benefit amounts to the nearest penny or to the nearest 10 cents as proposed by the Secretary of HEW. (See p. 8.)

--Eliminate the minimum benefit provision for new beneficiaries. To minimize the hardship to the few needy beneficiaries not eligible for SSI, the Congress could authorize a limited SSI payment which would replace the portion of the Social Security benefit lost through eliminating the minimum provision. (See p. 10.)

--Revoke section 224(d) of the Social Security Act, which allows States to offset their portion of disability benefits, and require that the Social Security offset

8

B-201375

be made effective when workers' compensation benefits are awarded, rather than when SSA is notified of the award. (See p. 12.)

--Determine SSI benefit eligibility and payment amounts on a monthly retrospective basis, rather than the quarterly prospective basis. (See p* 45.)

We also recommend that the Congress:

--Consider whether the Emergency Assistance program should continue. If the Congress determines that the program should continue, it should review the positions of HHS and the courts, including the U.S. Supreme Court, concerning eligibility and the type and extent of emergencies covered.' It should then, if necessary, amend the legislation to clearly indicate congressional intent. (See p. 59.)

--Determine, in cooperation with HHS, the controls that would best provide additional appropriate and feasible financial incentives and enact legislation to establish them to effectively control AFDC payment errors. The Congress should consider HHS' study to determine an ultimate' error rate goal in establishing the needed financial incentives. (See p. 61.)

Finally, we recommend that the House and Senate Appro- priations Committees retract the conference committee directive for Federal fiscal sanctions against the States, based on the AFDC quality control error rates. (See p. 61.)

- - - -

Copies of this report are being sent to the Director, Office of Management and Budget, and the Secretaries of Health and Human Services, Education, and State.

zz* ItA Comptro er GenLral of the United States

. . . . . .,A

., ‘! *.

.I’

Contents

APPENDIX

Paqe

I. ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING THE OLD-AGE AND SURVIVORS AND DISABILITY INSURANCE PROGRAMS

Social Security Student Benefits for Postsecondary Students Should Be Discontinued

Liberal Deposit Requirements of States' Social Security Contributions Adversely Affected Trust Funds

Savings to the Social Security System if Benefits Were Calculated to the Nearest Penny

Minimum Social Security Benefit: A Windfall That Should Be Eliminated

Legislation Authorizing States to Reduce Workers' Compensation Benefits Should Be Revoked

SSA Should Improve Its Recovery of Overpayments Made to Retirement, Survivors, and Disability Insurance Beneficiaries

Need for Stronger Action To Recover Overpayments to Beneficiaries Earning More Than the Allowable Amount

Problems in Detecting Duplicate Social Security Payments for Dependent Children

SSA Should Obtain and Use State Data to Verify Benefits for Its Programs

II. ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING THE SUPPLEMENTAL SECURITY INCOME PROGRAM

SSI Payment Errors Can Be Reduced Need To Prevent Windfall Benefits to

SSI Recipients Flaws in Controls Over the SSI Computerized

System Cause Millions in Erroneous Payments

Number of Newly Arrived Aliens Who Receive SSI Needs To Be Reduced

Need for Systematic Medical Review of Dis- abled SSI Recipients

1

1

6

8

10

12

14

17

18

20

23 23

25

27

30

34

::,

,‘-: .

;‘.

.

Page

APPENDIX

Erroneous SSI Payments Result from Prob- lems in Processing Changes in Recipients' Circumstances

SSA Should Improve Its Collection of Overpayments to SSI Recipients

Improvements Needed To Insure Accuracy of SSI Retroactive Payments

SSI Overpayments Due To SSA's Determining Eligibility and Benefits on a Prospec- tive Quarterly Basis

SSI Overpayments to Medicaid Nursing Home Residents Can Be Reduced

Need To Assess Penalties Against SSI Recipients Who Fail To Report Changes Affecting Eligibility and Benefit Payments

III. ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING THE AID TO FAMILIES WITH DEPENDENT CHILDREN AND EMERGENCY ASSISTANCE PROGRAMS

AFDC Overpayments Caused by Delays in Stopping Payments to Ineligible Recipients

Requirements and Practices for Refunding the Federal Share of Uncashed AFDC Checks

Need for Improved Overpayment Recovery Policies in the AFDC Program

State Advance Payments to AFDC Recipients Are Inconsistent with Federal Regulations

Should Emergency Assistance for Needy Families Be Continued? If so, Program Improvements Are Needed

Legislation Needed To Improve Program for Reducing Erroneous Welfare Payments

Better Management Information Can Be Obtained from the Quality Control System Used in the AFDC Program

AFDC Aid to Families with Dependent Children HEW Department of Health, Education, and Welfare HHS Department of Health and Human Services RSDI Retirement, Survivors, and Disability Insurance SSA Social Security Administration SSI Supplemental Security Income

ABBREVIATIONS

36

39

42

45

47

49

51

51

53

55

56

59

61

61

: !: 1’ ..G: ,I ‘,,

‘. ,‘L.,’

&. ,,. I I. I_ ::r - i ,. ..: .-

‘,. , ,*i

.;,. e., $1 :

APPENDIX I APPENDIX I

ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING THE

OLD-AGE AND SURVIVORS AND DISABILITY INSURANCE PROGRAMS ---

The Old-Age and Survivors Insurance program protects individuals and families from the risk of economic loss resulting from old age and death by providing income to retired workers, their dependents, and their dependent survivors. The program is financed by contributions to the Federal Old-Age and Survivors Insurance Trust Fund by employers, employees, and self-employed individuals based on earnings. In fiscal year 1981, it is expected that about 31 million beneficiaries will receive about $118 billion in benefits.

The Disability Insurance program protects individuals and families against the risk of economic loss resulting from disability by providing income to insured workers and their dependents who are unable to engage in substantial gainful activity because of medical impairment. This program is financed by contributions to the Federal Disability Insurance Trust Fund by workers, employers, and self-employed individuals based on earnings. In fiscal year 1981, it is expected that about 4.7 million beneficiaries will receive about $16.9 billion in benefits.

SOCIAL SECURITY STUDENT BENEFITS FOR POSTSECONDARY STUDENT% SHOULD BE DISCONTINUED I_.- (HRD-79-108, Aug. 30, 1979)

The basic purpose of the Old-Age and Survivors and Dis- ability Insurance programs is to provide some minimum family income in the event of the taxpayer's retirement, disability, or death. However, Social Security student benefits divert tax money from that basic purpose. During the 1979-80 school year, an average of about 796,000 students received benefits totaling an estimated $1.95 billion. Social Security student benefits also give many students more money than their school costs warrant and inequitably bar benefits to some students because they are part-time students or are married.

The Social Security Student Benefit program seemed appropriate when it was created in 1965. At that time there were only two Office of Education aid programs--the National Direct Student Loan and the College Work-Study programs. Since 1965 Office of Education aid programs have expanded to include four more programs --Basic Educational Opportunity Grants,

1


Guaranteed Student Loans, Supplemental Educational Opportunity Grants, and State Student Incentive Grants. These programs are now administered by the Department of Education--a new Executive Department created pursuant to Public Law 96-88.

We reported that the Office of Education was willing to provide aid more equitably to most postsecondary students now receiving payments from the Social Security Administration (%A). We reported that, if Social Security student benefits to postsecondary students were terminated effective fall 1980, the estimated net first-year savings would have been about $1.1 billion.

Another unfavorable aspect of financing student benefits through the Social Security trust funds is that such benefits are being financed through a regressive tax system. The So- cial Security tax, unlike the Federal income tax, does not tax higher earnings at higher rates. Also, those whose income from covered employment exceeds the maximum amount of annual earnings subject to Social Security tax would not contribute anything toward student benefits on that portion of their income exceeding the maximum amount.

On April 13, 1979, the Secretary of Health, Education, and Welfare (HEW) proposed legislation (the proposed Social Security Amendments of 1979) which provided for eliminating future awards of Social Security student benefits to postsecondary students. However, this proposed legislation was not enacted into law. On February 20, 1980, the Secretary of HEW proposed legislation (the proposed Social Security Amendments of 1980) which does not provide for eliminating future awards of Social Security student benefits. According to the Commissioner of Social Security, consideration of the proposal to eliminate student benefits and other proposed major changes in benefit structure has been deferred pending review and evaluation of the reports of the Advisory Council on Social Security, the 1979 study of the treatment of men and women under Social Security, and the Universal Coverage Study.

We recommended that the Congress enact an amendment to the Social Security Act which would discontinue student benefits for postsecondary students and take the necessary steps to assure that the Office of Education would have sufficient financial resources to meet any increased demand for its programs arising from discontinuance of these benefits.

2


Discontinuance of Social Security student benefits for postsecondary students could be accomplished in various ways. Following are ways to terminate the benefits with estimated dollar effects.

(billions) 1. Termination of benefits, effective

fall 1981: (The Department of Education has not compiled data on increased costs to the Basic Grant program from terminating Social Security student benefits beyond fiscal year 1982.)

Trust funds first-year savings $1.38

Increased cost to Basic Grant program

Net savings $1.12

.26

2. Termination of benefits through phaseout: (Benefits would not be payable to students in postsecondary schools who reach age 18 after August 1981 or were not getting student benefits before September 1981.)

Fiscal years 1982 1983 1984 1985 1986 Total - - P - -

(millions) Trust fund

savings $160 $527 $971 $1,474 $1,919 $5,051 Increased

cost to Basic Grant program 86 151 188 215 200 840

Net savings $2 $376 $783 $1,259 $1,719 $4,211

Trust funds savings are based on SSA data. The increased cost to the Basic Grant program is based on Department of Education estimates which indicate a need for paying former

3

‘, ii

,.,T . ( _. “ . ,

,y ; . ‘j 1

. , ” I

. : . , : : ( ,

. ”


student beneficiaries who have not been receiving basic grants and former student beneficiaries who have been receiving basic grants, but, because of discontinuance, would qualify for more basic grant money.

In addition to substantial savings, implementation of our recommendation should improve the overall administration of programs intended to aid students. Our longstanding position is that the consolidation of separate programs serving similar objectives (e.g., education aid programs) into broader purpose programs and the placement of programs serving similar goals within the same Federal agency (e.g., Department of Education) should increase the efficiency and effectiveness in the delivery and administration of Federal assistance.

As a result of information developed during our review, SSA, in March 1979, instituted a procedure to verify that student beneficiaries were attending school full time as required to be eligible for student benefits. Based on information provided by SSA, we estimated that this procedure will save about $36 million a year.

We also reported that, in July 1978, the Secretary of HEW testified that large numbers of Basic Grant applicants in school year 1977-78 may have wrongfully failed to report they were getting Social Security benefits. To curtail excess payments that arise from insufficient reporting, the Office of Education began to use a new computer procedure to match Basic Grant applicants against Social Security students. We tested the procedure on a sample of Social Security recipients of Office of Education Basic Grants in the 1977-78 school year and estimated that the Office of Education paid an excess of $23.8 million to Social Security recipients.

The procedure did detect student recipients receiving excess grants which accounted for about two-thirds or $15.6 million of the excess. It did not detect such nonstudent beneficiaries, as 17-year-olds and disabled workers, nor could it in the future unless the Office of Education matched against a larger number of Social Security recipient records.

Office of Education officials said that matching applicants against nonstudent beneficiaries would unduly delay award of grant eligibility. We then suggested a two-step alternative:

4


1. Before determining eligibility, match applicant records not only against the records of student beneficiaries, but also against the records of nonstudent beneficiaries in the 17- to 22-year-old age range.

2. After determining eligibility, send SSA a list of eligibles to be matched against the records of all Social Security beneficiaries--and have the matches produced by this procedure sent to the Office of Education for final verification of benefit amounts to avoid excess grants.

The first step would have detected $5.8 million of the $8.2 million that the procedure failed to detect in the 1977-78 test: the second step would have detected the other $2.4 million.

The first step of our recommendation was adopted, resulting in detecting and avoiding $5.8 million in grant overawards annually. However, the second step of our recommendation has not yet been implemented. A representative of the Department of Education said this proposed step is still being studied. As indicated above, we estimated that implementation of this step would reduce basic grant overawards by about $2.4 million annually.

Recommendation to the Congress

We recommend that the Congress amend the Social Security Act to discontinue student benefits for postseeondary students and take steps to assure that the Department of Education will have sufficient financial resources to meet any increased demand for aid arising from discontinuance of these benefits.

Recommendation to the Secretaries of Education and Health and Human Services

We recommend that the Secretaries require that the Social Security/Basic Grant computer matching procedure be revised to incorporate the second step of our recommendation, as previously discussed.

5

_’

; 1 >,.

: , ,

, / . : . *

. , I i.,

‘,

,


LIBERAL DEPOSIT REQUIREMENTS OF STATES' SOCIAL SECURITY CONTRIBUTIONS ADVERSELY AFFECTED TRUST FUNDS (HRD-79-14, Dec. 18, 1978)

Section 218 of the Social Security Act (42 U.S.C. 418) authorized voluntary agreements between HEW and the States under which the employees of Statestand their political subdivisions are/provided Federal Old-Age, Survivors, and Disability Insurance benefits under title II of the Social Security Act. Section 218(i) of the act requires that HEW's regulations for administration of voluntary agreements be de- signed to make the requirements imposed on States the same, so far as practicable, as those imposed on private employers. Notwithstanding this requirement, HEW was permitting the States to make quarterly deposits of Social Security taxes 1 month and 15 days after the end of each calendar quarter for covered State and local government employees even though private employers were required to deposit Social Security and withheld income taxes weekly, semimonthly, biweekly, or monthly--depending on the amount of taxes withheld. Further, HEW's deposit requirement was more lenient than the Internal Revenue Service's requirement. Most State agencies and local governments we visited were required to remit withheld income taxes to the Internal Revenue Service within 3 banking days after each quarter-monthly period in which a payday occurred (7th, 15th, 22d, and last day of the month).

The Social Security trust funds could have earned about $1.1 billion in additional interest from 1961-79 had States been required to deposit taxes more frequently--monthly instead of quarterly-- thus making the funds available to the trust funds for earlier investment. If the quarterly deposit requirements were continued, we estimated that about $1.2 billion in interest would be lost during the 5-year period 1980-84.

On March 30, 1978, HEW published in the Federal Register a proposed rule to require States to make monthly deposits of Social Security taxes 15 days after the end of each month (referred to as the 15-15-15 proposal). On November 20, 1978, HEW revised its proposal to require States to deposit taxes for each of the first 2 months of a calendar quarter by the 15th day after each month, but taxes for the third month of a quarter would not be due until 1 month and 15 days after the end of the third month (referred to as the 15-15-45 proposal).

6

,.’ . , I : ‘8.

, , ‘, . ..’

/’ . . - ,

“ ! . ‘.


The States' primary objection to more frequent deposits was the loss of interest earned and cash flow on Social Secu- rity taxes from the time employees are paid until deposits are made. We stated that any such financial assistance to the States should be specifically legislated and not provided at the expense of the Social Security trust funds. We recommended that the Secretary of HEW reconsider the decision to implement the 15-15-45 proposal, and we urged that semimonthly or biweekly deposits be required. At a minimum, we suggested that HEW's original 15-15-15 proposal would be a viable alternative.

On June 9, 1980, the Social Security Disability Amend- ments of 1980 (Public Law 96-265) were approved, which mandate a 30-30-30 requirement, i.e., States must deposit Social Secu- rity taxes within the 30-day period following the last day of each month. This requirement is more lenient than HEW's original (15-15-15) or revised (15-15-45) proposals. The Senate Committee on Finance report (S. Rept. 96-408) states that the 30-30-30 requirement was intended to ease the transition to HEW's 15-15-45 proposal. However, by enacting the 30-30-30 requirement into law, HHS will be precluded from making a transition into any other deposit requirement unless the Congress amends the law.

As shown below, changing the quarterly deposit requirement to the 30-30-30 requirement will result in an estimated $1.4 billion in additional interest revenues to the Social Security trust funds during fiscal years 1981-85. However, if as we urged, semimonthly or biweekly deposits were required and assuming that this requirement were effective beginning with fiscal year 1982, the trust funds could earn about $339 million more in interest than under the 30-30-30 requirement during fiscal years 1982-85.

7


Fiscal year

1981 $ - $ 154 $ - 1982 284 235 49 1983 367 286 81 1984 432 337 95 1985 526 412 114

Total $1,609 $1,424 $339

Estimated Additional Interest Income That Could Be Earned by the Trust Funds Over

Previous Deposit Requirements -4-

Difference Semimonthly/

Deposit requirements 30-30-30- Semimonthly 30-30-30-requirement requirement

(millions)


We recommend that the Congress amend the Social Security Act to require States to deposit Social Security taxes semimonthly or biweekly. We also recommend that the Congress consider requiring States to deposit Social Security taxes using the same schedule that States now use to deposit withheld income taxes. Such a requirement would enable the trust funds to earn additional interest income over the $339 million which could be earned by requiring remittances semimonthly or biweekly.

SAVINGS TO THE SOCIAL SECURITY SYSTEM IF BENEFITS WERE CALCULATED TO THE NEAREST PENNY (HRD-78-160, Sept. 8, 1978)

This report pointed out that section 215(g) of the Social Security Act (42 U.S.C. 415 (g)) requires Social Security benefit amounts which are not a multiple of $0.10 to be rounded to the next higher $0.10. We estimated that a savings of $386 million would accrue to the Retirement and Survivors Insurance program from calendar years 1980-86 if section 215(g) were amended to provide that benefits be calculated to the nearest penny. Our savings estimate considered only one adjustment each year for recipients' benefit amounts although some recipients' benefits are adjusted more frequently than once a year. Thus, our estimated savings should be considered

8


a conservative figure. A smaller savings would also be achieved for the Disability Insurance program. We recommended that section 215 (g) of the Social Security Act be amended to provide for computing benefit amounts to the nearest penny.

In April 1979, the Secretary of HEW proposed to the Congress a draft bill entitled the "Social Security Amend- ments of 1979." This draft bill included a provision to round benefit amounts to the nearest dollar, rather than the next higher 10 cents. However, this bill was not enacted into law.

In February 1980, the Secretary of HEW proposed to the Congress another draft bill entitled the "Social Security Amendments of 1980." This draft bill was later introduced as H.R. 6652, and includes a provision to round Social Secu- rity benefit amounts. However, unlike the 1979 proposed legislation, H.R. 6652 proposes to round benefit amounts to the nearest 10 cents rather than to the nearest dollar. The rounding proposal was changed because rounding to the nearest dollar would have caused the following problems.

--Distortion of the cost-of-living percentage increases. For example, if there is a lo-percent cost-of-living increase and one beneficiary is getting $74 and another is getting $75, neither beneficiary would get an increase of 10 percent. The first beneficiary would receive $81, for an increase of 9 percent, and the other would receive $83, for an increase of 11 percent. Thus, the difference caused by the rounding could be significant.

--In many instances two, three, or more roundings might be done in a benefit calculation. While in the aggre- gate there would be as many upward as downward roundings, it would be possible for all roundings to go in the same direction and beneficiaries might get benefits that are $3 or more higher or lower per month.

The estimated budgetary savings to the Social Security and SSI programs from rounding to the nearest 10 cents is as follows:

APPENDIX I APPENDIX ,I

Fiscal year Estimated savings

(millions)

1982 $ 8 1983 37 1984 62 1985 82 1986 100

Total 1982-86 $289 Z

Although the proposal to round benefit amounts to the nearest 10 cents is not the same as our recommendation to compute amounts to the nearest penny, the rounding proposal, if enacted into law, should substantially accomplish the same savings as our recommendation.

Recommendation to the Congress ----_-___-

We recommend that the Congress amend the Social Security Act to compute benefit amounts to the nearest penny or to the nearest 10 cents as proposed by the Secretary of HEW.

MINIMUM SOCIAL SECURITY BENEFIT: ~~i~~~~LTHAT-SHOUL~~~-ELIMINATED _---------------- (HRD-80-29, Dec. lo--ig?$)----------- -------4------ -- - -!-.. ----

We reported that the minimum benefit provision of the Social Security Act, intended to help the poor, had in recent years mainly benefited retired Government workers with pensions and homemakers supported by their spouses' incomes. The provision grants a much higher benefit than individuals have earned and would otherwise receive. For example, even though a worker's earned benefit is only $40 a month, he or she would receive the minimum benefit of $122 a month.

Contrary to Social Security's concept of partially re- placing a person's covered earnings upon retirement, the beneficiaries receiving the minimum benefit which we sampled received benefits that were about four times larger than their average monthly covered earnings. Most of the sampled minimum beneficiaries were part-time or intermittent workers--never a permanent part of the labor force covered by Social Security. On the average, the minimum beneficiaries had some work in covered employment in only about 1 of every 4 years. Nearly half had gaps in employment of 20 or more years.

10


Sampled minimum beneficiaries generally could not have depended primarily on their earnings from covered employment because they were too low. Their average covered earnings were only about $22 a month for the period 1953-76. Only 3 percent had earned as much as $4,000 during any single year in that time period, and only one-third had earned as much as $2,000 in any one of those years.

In 1974 the Supplemental Security Income (SSI) program was implemented. This program established a Federal minimum income level for needy people who are at least age 65 or blind or disabled. Before the program, the minimum Social Security benefit may have been the only source of income for many people, but now most needy elderly are eligible for SSI.

In the Social Security Amendments of 1977, the Congress froze the entry level of minimum beneficiaries at $122 as of January 1979 because of a growing concern that the benefit was a windfall to people who had not worked regularly under Social Security. The minimum was not eliminated for fear a sharp drop in the benefit level might cause hardships for needy people. Also, the Congress believed that freezing the minimum would, with inflation, result in a gradual elimina- tion of the minimum. According to SSA, it will take more than 30 years for the freezing action to eliminate minimum benefits under the current law.

On April 13, 1979, the Secretary of HEW proposed legislation (the proposed Social Security Amendments of 1979) which provided for eliminating the minimum benefit for new beneficiaries. In our report, we recommended that the Con- gress approve the proposal. To minimize the hardship of the few needy beneficiaries who would not be eligible for SSI, we recommended that the Congress consider authorizing a limited SSI payment which would replace the portion of the Social Security benefit lost when the minimum provision is eliminated. However, the proposed legislation was not enacted.

On February 20, 1980, the Secretary of HEW proposed legislation (the proposed Social Security Amendments of 1980). However, this proposal does not provide for eliminating the minimum benefit for new beneficiaries as did the 1979 legislative proposal. According to a Department representative, consideration of the proposal to eliminate the minimum benefit provision and other proposed major changes in benefit structure has been deferred pending review and evaluation of

11

. , , I I

, . . . , I . . : - :

: ,,‘I “.’ ./‘; 1’

/ “ : : . . ,<, , : I

I ’ - , ,_,‘.. ’ : “ . “ ;

. . J.. “3

; . : . : .


the reports of the Advisory Council on Social Security., the 1979 study of the treatment of men and women under Social Security, and the Universal Coverage Study.

As shown below, eliminating the minimum benefit for new beneficiaries beginning in fiscal year 1982 would save an estimated $405 million during fiscal years 1982-86.

Fiscal years 1982 1983 1984 1985 1986 "Total - - P --

(millions) Savings to

Social Security $50 $95 $135 $165 $205 $650

Increases in SSI 15 35 55 65 75 245 - -

Net savings $g $60 $80 $100 $130 $405


We recommend that the Congress eliminate the minimum benefit provision for new beneficiaries. To minimize the hardship to the few needy beneficiaries not eligible for SSI, the Congress could authorize a limited SSI payment which would replace the portion of the Social Security benefit lost through eliminating the minimum benefit provision.

LEGISLATION AUTHORIZING STATES TO REDUCE WORKERK'COMPENSATION BENEFITS SHOULD BE REVOKED (HRD-80-31, Mar. 6, 1980)

In 1965 the Congress added an offset provision to the Social Security Act to limit combined benefits when workers receive both State workers' compensation and Social Security disability benefits. This provision was intended to provide disabled workers with a financial incentive to return to work. Before the 1965 provision, benefits under the two programs could exceed the amount a worker was earning before becoming disabled.

Under the offset provision, a worker's Social Security disability benefits are to be reduced so that the combined payments from both programs do not exceed a specified amount to be determined according to a formula specified in the law.

12


However, the offset provision states that the workers' compensation offset will not be applied by SSA if a State law or plan provides for an offset of workers' compensation benefits.

We reported that the provision allowing States to offset or reduce workers' compensation benefits can shift some of the financial responsibility for work-related disabilities from employers (who bear the cost of workers' compensation benefits) to employees (who bear some of the cost of disability benefits). We reported also that State offsets reduced the savings SSA could achieve through offsetting and in some cases resulted in combined benefits which exceeded the maximum benefit amount determined pursuant to the formula in the law.

During our review, 11 States had offsetting laws or plans which were considered adequate to preclude Federal offsets and were offsetting their workers' compensation benefits. We reported that SSA's offsets were saving the Social Security Trust Fund about $147 million annually and SSA's costs of administering the offset provision were about $2.5 to $3 million, leaving a net savings of about $144 million annually. Also, we reported that if all States made the offset, the Social Security Trust Fund could lose $160 million annually by 1981.

In addition, we reported that the language of the offset provision was not precise as to whether SSA could apply its offset retroactively when recipients did not accurately and promptly report their workers' compensation benefits. In 1976, after getting an opinion from HEW's Office of General Counsel, SSA adopted a policy of offsetting for workers' compensation benefits only in the months after receiving notification of entitlement from the disabled worker. As a result, disabled workers who fail to report their workers' compensation benefits promptly or accurately can receive excessive benefits.

We recommended that the Congress amend the Social Security Act to:

--Revoke section 224(d), which allows States to offset their portion of disability benefits.

--Require that the Social Security offset be made effective when workers' compensation benefits are

13


awarded, rather than when SSA is notified of the award.

SSA officials told us that two States--New Jersey and Wisconsin --had recently passed legisla+ion providing for State offsets which should be operational by the end of 1980. Using data provided by SSA, we estimate that the Social Security Trust Fund could lose as much as $5 million annually as a result of these two additional State offsets. We were advised also that other States are considering offset programs which, if adopted, would further increase trust fund losses.

Recommendations to the Congress

To prevent further losses to the Social Security Trust Fund and reduce the potential for excessive payments, we recommend that the Congress ,lmend the Social Security Act to:

--Revoke section 224(d), which allows States to offset their portion of disability benefits.

---Require that the Social Security offset be made effective when workers' compensation benefits are awarded, rather than when SSA is notified of the award.

SSA SHOULD IMPROVE ITS RECOVERY OF OVERPAYMENTS MADE TO RETIREMENT, SURVIVORS, AND DISABILITY INSURANCE BENEFICIARIES (HRD-79-31, Jan. 17, 1 979)

During January to July 1978, SSA identified overpayments of more than $536 million to 967,000 individuals. Most of these overpayments were made to persons eligible for future benefits --allowing SSA to recoup these overpayments by ad- justing current and future benefits. A sizable portion of this amount will be recovered by this method. However, we reported that individuals no longer receiving benefits pre- sented a difficult recovery problem. These individuals owed SSA about $234 million as of July 31, 1978.

Our report discussed several management problems SSA was experiencing in collecting these overpayments, including:

--SSA's recovery personnel were making questionable and erroneous decisions resulting in monetary losses.

14

_/I

. . : . i

.‘. ,.‘.

‘; , , . , :

l .:’ .‘. : - ”

).’ : .

‘_ I .

. . .


--SSA's management information system did not sufficiently provide the type of information needed by managers to evaluate recovery efforts.

--Some SSA policies and procedures governing recovery were not being properly and consistently applied.

We reported that, until recently, SSA headquarters management had been slow to react to its mounting overpayment and recovery problems, and its approach to recovery had lacked overall direction.

The following summarizes our recommendations to correct management problems and SSA's action in response to those recommendations.

Recommendation: Determine to what extent full-time specialists in district offices could do a more effective job recovering overpayments from beneficiaries and recipients.

SSA action: SSA made an experiment in a district office to test the efficiency of centralizing the overpayment recovery workload and determine if a full-time overpayment specialist position is warranted. The experiment is completed, but SSA has not completed its analysis of the results.

Recommendation: Require managerial personnel at district offices to provide service representatives with greater direction, supervision, and feedback on their recovery work.

SSA action: In April 1979 instructions were sent to SSA regional commissioners which requested area direc- tors to make indepth reviews during July to September 1979 of field offices' overpayment processing. The primary purpose of these reviews was to insure that district management is actively involved in overpayment workload processing and in increasing the prompt- ness and consistency of overpayment dispositions.

Recommendation: Immediately refine a management information system to define the exact composition of the outstanding balance on unsettled accounts. This should include potential adjustment cases, accounts being recovered through installments, cases where recovery will

15


be attempted from individuals no longer on the benefit rolls, and the length of time each overpayment has been outstanding.

SSA action: SSA has not refined the management information system. According to SSA, lack of staff and changing priorities have delayed their effort to implement this recommendation.

Recommendation: Assure that the SSA task force report on recovery reviewer alternatives, issued in August 1978, re- ceives immediate attention. Further appropriate action should be taken to provide recovery reviewers with feedback on their work, and efforts must be made to overcome the present inadequate technical leadership, training, and quality assessment at the program service centers.

SSA actions: SSA stated that the task force recommendations are being implemented. For instance, as recommended by the task force, the duties of claims authorizer technical assistants in program service centers have been revised to require them to provide quality assessment and technical assistance to recovery reviewers.

Recommendation: Examine recovery provisions of the c'l~ims manual to clarify procedures, eliminate inconsistencies, and provide more explicit direction in terms of direct contact with the overpaid individual and who should do it, negotiation of compromise offers, and evaluation of financial data.

SSA actions: Revisions to the claims manual were issued in July 1979 which address most of these issues.

Recommendation: Reevaluate the computer program provision, which authorizes automatic termination of any account that is or drops below $200 after one demand notice is issued.

SSA action: SSA plans to change the computer program provision to provide that three instead of one followup refund requests will be mailed to overpaid recipients whenever an overpayment is or drops below $200. How- ever, this change has not yet been implemented.

16


Recommendations to the Secretary of Health and Human Services

We recommend that the Secretary direct the Commissioner of SSA to:

--Complete the analysis of the district office experiment with centralizing overpayment recovery workload and the need for a full-time overpayment specialist.

--Immediately refine its management information system to define the exact composition of the outstanding balance on unsettled accounts. This should include potential adjustment cases, accounts being recovered through installments, cases where recovery will be attempted from individuals no longer on the benefit rolls, and the length of time each overpayment has been outstanding.

--Change the computer program provision, so that termination will not automatically occur on any account that is or drops below $200 until after three demand notices are issued.

NEED FOR STRONGER ACTION TO RECOVER OVERPAYMENTS TO BENEFICIARIES EARNING MORE THAN THE ALLOWABLE AMOUNT (HRD-79-89, July 2, 1979)

The Social Security Act, section 203(b),(f), as amended, requires that persons under age 72 receiving Social Security benefits have their benefits reduced if they earn income over an exempt amount. SSA requires Social Security beneficiaries who continue to work to estimate their current year's earnings; benefits are reduced if the estimated earnings exceed the exempt amount. Earnings estimates may be changed any time during the year, and SSA adjusts the individual's benefits accordingly. After the close of the earnings year, beneficiaries whose earnings exceed the exempt amount must file an annual report of earnings with SSA by April 15. The annual report form also includes a place to estimate the following year's earnings.

If the annual reported earnings differ from estimated earnings, SSA adjusts benefits to recoup overpayments or pay underpayments. Also, every employer of persons covered by Social Security is required to report the amount of wages

17


paid each employee. SSA makes an automated comparison of earnings information from the employer's report with earnings reported by the beneficiary. An earnings enforcement case is generated when the comparison shows

--the employer(s) reported wages paid to a beneficiary that exceed the allowable amount but the beneficiary did not file the annual report, or

--a beneficiary reports annual earnings different from those reported by the employer(s).

We reported that during 1974-76, SSA failed to follow through on an estimated uncleared 83,000 enforcement cases involving about $39 million in potential overpayments and $5 million in potential underpayments. An additional $8.9 million in potential overpayments went undetected because of SSA's practices that disregard enforcement cases involving terminated student beneficiaries. Overall, the Social Secu- rity Trust Funds could lose about $43 million because SSA did not take proper action on these earnings enforcement cases.

We recommended that the Secretary of HEW monitor the efforts of the Social security Commissioner to resolve all uncleared 1974-77 earnings enforcement cases and improve the control system so that earnings enforcement cases continue to be periodically called up until they are resolved.

In response to our recommendations, SSA began to resolve the enforcement cases we identified and recovered about $16 million in overpayments and, penalties. SSA also improved its control system for earnings enforcement cases which should result in recurring savings of about $8.7 million annually.

PROBLEMS IN DETECTING DUPLICATE SOCIAL SECURITY PAYMENTS FOR DEPENDENT CHILDREN (HRD-79-27, Dec. 22, 1978)

We found that SSA made 329 duplicate payments of student benefits in May 1977. Of these duplicate payments, SSA identified 99 but did not detect the other 230. Our calculation of the duplicate payments for the 230 students was $616,000 for all months duplicate student payments were made through June 1978. If this calculation is typical of all dependent children for whom benefits were being paid in

18


May 1977, the total undetected duplicate payments could amount to about $4.2 million for all months duplicate payments were made through June 1978. These payments were not discovered because SSA's daily detection system was not fully operational, and its annual duplicate detection system criteria were too restrictive.

We also reported that SSA had identified but not resolved about 1,900 cases of potential overpayments. We estimated that these cases could involve $1.9 million, based on SSA's overpayment experience with the types of cases that it already reviewed.

Finally, we identified about 77,000 (9 percent) students who did not have a Social Security number recorded in the payment system. Without accurate Social Security numbers, SSA cannot independently apply the earnings test to those students who have not reported such earnings and possibly reduce or withhold benefits.

We recommended that the Secretary of HEW direct the Commissioner of SSA to:

1. Determine from other existing social security records the social security numbers for those dependent children missing their numbers, especially students, and record them in the payment records.

2. Compare the social security numbers of all dependent children currently receiving benefits to eliminate duplicate payments or to correct instances where different dependents have the same recorded social security number.

3. Change SSA's duplicate payment detection system to correct the type of problems disclosed by our review which caused the estimated $4.2 million overpayments.

4. Assure that the potential duplicate payments which are identified by SSA's duplicate payment operations are reviewed and corrected in a timely manner.

SSA is developing a pilot project to determine the feasibility of obtaining missing social security numbers from other SSA sources as we recommended. However, this effort is not complete. Also, SSA has not acted in response to our second recommendation.

19

APPENDIX I APPENDIX, I

SSA has modified the duplicate payment detection system to increase its capabilities to identify and correct overpayments. This action will result in undetermined savings in future years. SSA has also taken action to assure that potential duplicate payments identified by the system are examined and corrected. The $1.9 million in unresolved potential duplicate payments have now been resolved.


We recommend that the Secretary:

--Monitor SSA's efforts to determine from other existing social security records the social security numbers for those dependent children missing their numbers, especially students, and record them in the payment records.

--Direct the Commissioner of SSA to compare the social security numbers of all dependent children currently receiving benefits to eliminate duplicate payments or to correct instances where different dependents have the same recorded social security number.

SSA SHOULD OBTAIN AND USE STATE DATA TO VERIFY BENEFITS FOR ITS PROGRAMS -(HRD-80-4, Oct. 10, 1979)

SSA administers four programs that pay monthly benefits to about 39.5 million aged and disabled persons and/or their survivors and dependents. The integrity of these programs depends, to a large degree, on recipients voluntarily reporting changes in their income, resources, or other circumstances which can affect their eligibility or benefits. Benefits under one or more SSA programs can be affected by several State and local government programs, such as unemployment compensation, workmen's compensation, State pensions, and general assistance. In addition, most States maintain information on such things as births, marriages, and deaths.

We reported that SSA had obtained and verified benefit information from other Federal agencies to help minimize incorrect payments, but little had been done to obtain information from State and local governments which could be used to further reduce erroneous payments. We reported that SSA's


efforts to obtain State and local records, for the most part, had been fragmented and uncoordinated. We reported also that the work we did with California and New York unemployment insurance records showed that about $1.6 million in erroneous SSI benefits could have been detected annually in these two States through a data exchange.

We recommended that SSA be directed to develop and implement a comprehensive national effort to obtain and use State and local data, noting, where appropriate, legislative and administrative impediments to obtaining such data. Significant impediments should be brorlght to the attention of the Congress and/or HEW for resolution. We recommended also that SSA be directed to request the assistance of California and New York in obtaining unemployment compensation records and use these records along with current Pennsylvania and Kentucky workmen's compensation data to verify SSA's SSI and disability insurance records.

In August 1980 SSA reported that it was gradually expanding the use of State data to verify eligibility in all its income maintenance programs. It reported, however, that avail- ability of resources, implementation of the Disability Amend- ments of 1980, and other mandatory systems' tasks had left little time to make a comprehensive national effort to obtain and use State and local government data to verify benefits for its programs (interfaces). Instead, SSA continued its phased program, State-by-State approach which it believes will prove more cost effective.

SSA also said that it was completing a comprehensive survey of Federal/State data exchanges in four States and that the survey had produced valuable inforn&ation on how future interfaces might work in the aid to families with dependent children (AFDC), SSI, and retirement, survivors and disability insurance (RSDI) programs. An SSA official advised us in October 1980 that a survey report was being prepared which would contain recommendations on how to proceed with large- scale program interfaces, including any relief that may be needed for administrative or legislative impediments.

In addition, SSA reported that it was developing a State and Federal exchange program to expand on State use of SSI records. SSA officials told us in October 1980 that under this program they were working with New York to exchange data for an SSI-State unemployment compensation interface. They told us also that SSA hopes to run the first interface early in 1981.

21


We were also advised that unsuccessful attempts were made to implement a similar data exchange in California. Instead, the State has agreed to consider an SSI-State public employee and teacher pension interface. SSA officials said that a formal request for this interface was made in Septem- ber 1980. In addition, Mississippi has recently agreed to an SSI-State AFDC data exchange. The data exchange will ini- tially be made for two counties and, if cost effective, will be expanded to the entire State. SSA officials stated that the two-county interface should be completed early in 1981 and should result in large savings for both the Federal and State governments.

SSA reported also that it had completed the interface of its SSI and disability programs with Pennsylvania's workmen's compensation program and was evaluating the results. In October 1980, we were advised that the evaluation had not been completed. In addition, SSA has requested workmen's compensation data from the State of Kentucky for an SSI and disability program interface. This request was made in the latter part of September 1980.

SSA is also participating in a pilot interface with a private insurance company, involving workmen's compensation records for New York State. SSA officials told us in October 1980 that data on about 1,000 cases had been received and matched against SSA records, but that the final results of these matches were not yet available.

SSA's actions fall short of the comprehensive national effort recommended in our report. However, we recognize that there are obstacles confronting SSA in obtaining State data and that some progress has been achieved in obtaining and using State and local government data to compute benefits under SSA's programs.

Recommendation to the Secretary -------r--- of Health and Human Services

Because of the large cost savings potential through more effective data exchanges., we recommend that the Secretary monitor SSA's efforts to develop and implement a comprehensive program to obtain and use State and local data to compute benefits for all its income maintenance programs.

22

APPENDIX II APPENDIX II

ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING ---we

THE SUPPLEMENTAL SECURITY INCOME PROGKAlll ---.---_--.-__.--_------ -_--_--- _--- - - - _ -

Title XVI of the Social Security Act established the Federal Supplemental Security Income program for the aged, blind, and disabled. In 1980 the SSI program provided a minimum income of $208.20 per month for an eligible individual and $312.30 per month for an eligible couple. Each July, SSI benefit levels are increased by the same cost-of-living percentage as Social Security benefits.

SSI is administered by the Federal Government under national uniform eligibility requirements and payment support levels. States are required to supplement the Federal benefit to assure that recipients of benefits under the former State- administered programs suffered no loss of income under SSI. States may make additional supplementary payments and may enter into agreements with the Federal Government to administer their supplementary payments. The Federal cost of administering the States' supplementary payments is financed from Federal funds. During fiscal year 1981, it is expected that about $6.1 billion will be paid to about 3.7 million aged, blind, and disabled recipients.

SSI PAYMENT ERRORS CAN BE REDUCED (HRD-7Gli5g,w 'tid;; ..is',- ‘ig78 ,- - - - - - - - - - - - - - - - - . - - - - - - A - - - - - .

The Social Security Amendments of 1972 established the SSI program which provides cash assistance to needy, aged, blind, and disabled persons based on nationally uniform eligibility requirements and benefits criteria.

SSA considers all income and resources of an SSI applicant in computing benefits. The applicant provides SSA with this and the other information for an eligibility determina- tion. Once eligibility is established, the recipient must report any subsequent change in income, resources, or other circumstances which would change a payment amount or affect eligibility. If any of this information is incorrectly reported or reported later than required, an overpayment or underpayment may result.

Compensation and pension benefits paid by such agencies as the Office of Personnel Management, Railroad Retirement Board, and Veterans Administration are required to be reported by SSI applicants. SSA requests that information on these benefits be reported when a recipient first applies

23


for SSI and when a recipient's case is reviewed for continued eligibility.

We reported that most of the data on Railroad Retirement Board and Veterans Administration benefits used in calculating SSI payments were inaccurate and estimated that use of more accurate data obtained from these agencies would reduce SSI overpayments by about $60 million a year and underpayments by about $4 million a year. We also estimated that about 36,000 ineligible recipients would be removed from SSI's rolls, resulting in further savings in administrative and medical assistance costs.

We recommended that SSA obtain complete and accurate compensation and pension data on a timely and continuing basis from the Railroad Retirement Board and Veterans Admin- istration for computing SSI payments. We also recommended that SSA review other Federal benefit payments to determine the need for and the feasibility of obtaining benefit information directly from other agencies.

In response to our recommendation, SSA now obtains pension and compensation data from the Department of Labor, Railroad Retirement Board, Veterans Administration, and Office of Personnel Management and interfaces these data with that reported by applicants and recipients to assure that SSI payments are correct. SSA is also continuing to explore the need for and feasibility of obtaining compensation and pension data from other Federal agencies.

SSA completed its first Veterans Administration interface in September 1976 and began making quarterly interfaces in 1977. These interfaces are currently made five times a year and SSA officials told us in July 1980 that efforts were underway to have a continuous data exchange whereby changes will be acted upon as soon as the information becomes available. The initial interface with the Department of Labor was completed in February 1976, the Railroad Retirement Board in January 1977, and the Office of Personnel Management in June 1978. Department of Labor and Railroad Retirement Board interfaces are made annually and Office of Personnel Manage- ment interfaces are made semiannually. SSA officials told us that they were working with the Department of Defense to obtain information on military retirement benefits, but that priority is being given to developing continuous interfaces with the Railroad Retirement Board and the Veterans Admin- istration because of greater savings potential.

24


Savings resulting from reductions in SSI payments will continue until the recipient dies or stops receiving SSI for reasons other than interface adjustments. Thus, the savings resulting from interfaces will generally continue for years. We estimate that, since SSA began its interfaces through fiscal year 1980, the total reduction in SSI payments was about $267 million. We were advised by SSA officials that about 60 percent of the savings would accrue to the Federal Government and about 40 percent to the States. Therefore, about $160 million of the total reduction accrued to the Federal Government and about $107 million accrued to the States.

We project that during fiscal year 1982, reductions in SSI benefits will be about $102 million. About $61 million will accrue to the Federal Government and about $41 million will accrue to the States. Our projections are based on the reductions in SSI payments achieved by the interfaces completed with the Railroad Retirement Board, Veterans Adminis- tration, and Office of Personnel Management through fiscal year 1980 and projected reductions for fiscal year 1981.

NEED TO PREVENT WINDFALL ------ BENEFITS TO SSI RECIPIENTS ---____-.- (HRD-80-44, May_30, 1980) ----- --I~

We reported that, in fiscal year 1977, SSI recipients received windfall benefits totaling an estimated $43.6 million as a result of receiving retroactive Social Security retirement, survivors, or disability benefits. Also, SSI recipients received undetermined windfall benefits as a result of receiving income from other sources, including veterans' compensation and pensions, and railroad retirement benefits.

SSA is required to determine SSI eligibility and benefit payment amounts on a quarterly basis. It computes benefits prospectively: that is, benefits are based on the income a recipient expects to receive over a projected 3-month period. Basic benefits are reduced dollar-for-dollar for countable income. Anything of value received by the individual or couple, including retroactive benefits received from sources other than SSI, is included as countable income, except for certain excluded amounts. Once computed, SSI payments are disbursed in equal monthly installments.

25


Windfall benefits occur when other retroactive income covering prior quarters is received in a current quarter and is greater than the current quarter's SSI benefits. Because the Social Security Act does not provide for the recovery of these windfall payments, a program inequity is created allowing recipients of large retroactive payments to receive more SSI benefits than recipients who receive a similar amount of non-SSI income on a monthly basis.

In our report, we recommended that the Congress give favorable consideration to proposed legislation which would allow the offset of SSI windfall benefits against retroactive Social Security payments. On June 9, 1980, Public Law 96-265 was enacted. Section 501 of the law provides for adjustment of retroactive title II Social Security benefits under the Old-Age, Survivors, and Disability Insurance programs on account of SSI benefits. This section is effective for title II benefits, entitlement for which is determined on or after July 1, 1981. The estimated savings which will result from this provision are as follows:

Fiscal year

Reduction of Additional windfall administrative Net benefits cost savings

(millions)

1981 $ 1 $2 S(l) 1982 27 3 24 1983 30 30 1984 33 33 1985 35 35 -

Total 1981-85 $126 X $121 -

We also recommended that after the proposed legislation was enacted, the Secretary of HHS direct the Commissioner of Social Security to

--take the necessary administrative action to change processing of concurrently filed claims to reduce the number of windfall benefits paid to SSI recipients receiving retroactive Social Security payments before retroactive SSI payments and

--advise the Congress whether additional legislation is needed to eliminate all windfall benefits to (1) SSI recipients who receive retroactive Social Security

26


benefits before retroactive SSI benefits and (2) SSI recipients who receive retroactive benefits from other Federal benefit-paying programs, such as the Veterans Administration or the Railroad Retirement Board.

To implement our recommendation, HHS stated that it would examine the feasibility of sequencing SSI and Social Security Title II benefits so that SSI retroactive payments will be made first whenever such payments are imminent. An SSA representative informed us on November 10, 1980, that this exami- nation was underway. Also HHS said that, when the offset provision (Section 501 of Public Law 96-265) becomes operational in July 1981, it would consider the need for additional Social Security legislation and determine the extent of SSI windfalls caused by retroactive benefits from programs of other Federal agencies. HHS said also that when this information becomes available, it will have to be assessed by the other Federal agencies, since any proposed remedial legislation would affect their programs.

FLAWS IN CONTROLS OVER THE SSI COMPUTERIZED SYSTEM CAUSE MILLIONS IN ERRONEOUS PAYMENTS (HRD-79-104, Aug. 9, 1979)

We reported that internal control weaknesses over SSA's complex computer system have resulted in over $25 million in erroneous benefit payments to SSI recipients, some of which occurred as early as January 1974. Based on recipient records existing as of September 1978, we estimated that about $20 million of the erroneous payments have occurred in the SSI program because of inadequate controls in the automated data exchange between the Retirement, Survivors, and Dis- ability Insurance and SSI computerized systems. We also estimated that about $5.4 million of the erroneous payments have occurred because of inadequate controls over the process by which field office personnel manually calculate benefit payment amounts.

We made 11 recommendations to the Secretary of HEW for improving the controls over the SSI computerized system. As of August 1980, SSA had taken action on two recommendations and had begun to implement, at least in part, most of the other nine recommendations.

27

APPENDIX II

Imolemented recommendations

APPENDIX II

SSA had taken action in response to the following two recommendations.

We recommended that SSA modify the SSI computerized system to properly post RSDI eligibility decisions to all appropriate data segments in the SSI computer master record. SSA stated that a systems modification was implemented in December 1979 which should correct this deficiency.

We recommended that SSA establish more controls over forced payment cases (cases in which payment amounts are manually calculated by field office personnel), assuring that all posteligibility events affecting these cases are processed in a timely manner and that these cases are returned to regular payment status as soon as possible. Field office procedures were updated in March 1980 to strengthen controls over forced payment cases. The new procedures also require that a district or branch office operations super- visor or higher official must approve the use of forced payment input by confirming that a defined systems limitation exists. These new procedures should help resolve the weak- ness which we estimate had caused $5.4 million erroneous payments ($3.4 million in overpayments and $2 million in underpayments),.

Recommendations not fully implemented

SSA had begun to implement, at least in part, most of the following nine recommendations. However, SSA's actions are either incomplete or promised actions are not adequate to fully resolve the problems identified. SSA should:

--Correct deficient exception controls in the SSI system, especially for such items as income and resources, which directly affect program eligibility and benefit payment amounts.

--Improve the documentation of the system's exception control process at the field office level and maintain up-to-date consistency between actual programed excep- tions and supporting documentation.

--Restrict the system override capability to supervisory personnel who have the appropriate authority to make these override decisions and to enter them into the computer system.

28


--Remove the data exchange override capability and "default on verification" provision from the SSI computerized system. We estimated that inappropriate use of these provisions had caused erroneous SSI payments of about $6.4 million.

--Modify the RSDI computer system to provide a complete payment history to the SSI system. We estimated that $6.3 million of erroneous payments ($6.1 million overpayments and $.2 million underpayments) occurred because a complete history of RSDI benefit payments was not entered, verified, and used for calculating SSI eligibility and benefit payment amounts.

--Determine why field office personnel do not enter all eligibility decisions into the RSDI computer system and take appropriate corrective action to ensure that these data are exchanged with the SSI computerized system. We estimated that over $7.2 million of erroneous SSI payments were made to applicants who file concurrent claims for both SSI and RSDI benefits because applicants' RSDI benefit amounts were not communicated and properly posted to the SSI computerized system.

--Modify the SSI system to exchange additional data ele- ments, such as recipient address and household composition, with the RSDI system to reduce the potential for erroneous payments and program fraud and abuse.

--Remove, where applicable, the system limitations that necessitate the manual calculation and control of forced payment cases.

--Review existing forced payment cases to (1) identify the reasons for forced payments, (2) verify the accuracy of all payments made, and (3) return cases not required to be force paid to regular payment status as soon as possible.

Recommendation to the Secretary of Health and Human Services

We recommend that the Secretary direct the Commissioner of Social Security to fully implement the nine recommendations not fully implemented.

29

,

1


NUMBER OF NEWLY ARRIVED ALIENS WHO RECEIVE SSI NEEDS TO BE REDUCED (HRD-78-50, Feb. 22, 1978)

We reported that about $56 million in SSI was provided annually to newly arrived aliens (other than refugees) in five States. In most cases, this and other public assistance was supplied because aliens' sponsors failed to keep their promises of providing support. We made recommendations to the Congress and to the Secretaries of State and HEW to correct the problems identified.

We recommended that the Congress enact legislation:

--Establishing a residence requirement to prevent assistance payments to newly arrived aliens, if the condition upon which eligibility is established existed before entry.

--Making the affidavit of support legally binding on the sponsor.

--Making aliens subject to deportation if they receive Federal, State, or local public assistance because of conditions existing before entry by defining public charge to include receiving any public assistance, regardless of whether repayment is required.

The Congress did not adopt these recommendations, but did take action which should result in fewer newly arrived aliens receiving SSI. On June 9, 1,980, Public Law 96-265 was enacted which provides that, for the purposes of eligibility for SSI benefits after September 30, 1980, aliens will be deemed to have the income and resources of their sponsors available for their support for a period of 3 years after their entry into the United States, unless the alien becomes blind or disabled after entry. The law provides specific requirements for determining the amount of sponsors' income and resources to be deemed available to aliens for determining eligibility for SSI.

During the 3 years after entry into the United States, an alien may be eligible for SSI benefits only if his sponsor agrees to and does provide such information as the Secretary of HHS requires to carry out the law's provisions. The alien and sponsor are jointly and severably liable to repay any SSI benefits incorrectly paid because of the sponsor's providing misinformation or because of his failure to report correct

30


information, except where the sponsor was without fault, or where good cause for such failure existed. Overpayments may be withheld from any subsequent payments under the Social Security Act to the alien or his sponsor. The estimated reductions in SSI payments resulting from this legislation are shown below.

Fiscal years 1981 1982 1983 1984 1985 Total ~ - - - -

(millions)

SSI $4 $12 $19 $24 $28 $87

We believe that the provisions of Public law 96-265 provide necessary statutory authority for substantially reducing the number of newly arrived aliens who receive SSI. Our review showed that in most cases, SSI and other public assistance was supplied because aliens' sponsors failed to keep their promises of providing support. Also, data obtained by HHS in connection with its alien overseas asset study show that the average amount of annual income alleged by sponsors was $18,800 and average alleged sponsor resources were $40,000. The data also indicate that, if sponsor income and resources were "deemed" to aliens, most of the newly arrived aliens would not have been eligible for SSI.

Although the provisions of Public Law 96-265 should contribute toward substantial reductions in the number of newly arrived aliens receiving SSI, we believe it is unlikely that this legislation will completely resolve the problems discussed in our report. For example, Public Law 96-265 is only applicable to aliens applying for SSI after September 30, 1980. Thus, aliens who applied before that date may still qualify for SSI without regard to the restrictions imposed by Public Law 96-265. Also, if aliens are sponsored by persons whose income and resources are below the amounts determined pursuant to the requirements contained in Public Law 96-265, those aliens may still qualify for SSI. As shown below, during fiscal years 1981-85, an estimated $118 million in SSI benefits will continue to be paid to newly arrived aliens (those arriving within the previous 3 years) unless further action is taken to limit these payments.

APPENDIX II

Fiscal year

APPENDIX II

Amount

(millions)

1981 $ 32 1982 32 1983 24 1984 15 1985 15

Total $118

We recommended that the Secretary of State emphasize to consular officers the importance of screening aliens who may apply for public assistance. On December 11, 1979, the De- partment of State sent an airgram to most of its diplomatic and consular posts emphasizing the importance of screening aliens, as we recommended.

We also recommended that the Secretary of State, in cooperation with the Secretary of HEW, develop more stringent income criteria for judging the ability of a sponsor to support a visa applicant. In commenting on our report in May 1978, the Department of State did not concur with this recommendation and said that aged parents seeking to join their children in the United States would be the aliens most adversely affected. The Department said this would be dis- criminatory against the elderly and would violate the reuni- fication of family concepts inherent in the immigration laws.

However, a Department of State representative informed us in September 1980 that consideration was being given to adopt- ing the "deeming" formula established by Public Law 96-265 for determining aliens' eligibility for SSI and prescribing this formula as a general guide for consular officers to use in determining whether aged, blind, or disabled visa applicants are likely to become public charges and thus ineligible for admission into the United States. The proposed use of the "deeming" formula as a criteria for determining visa applicants eligibility for entry into the United States has potential to reduce the number of aliens gaining admission to the United States who would be eligible for SSI.

Finally, we recommended that the Secretary of HEW direct the Commissioner of Social Security to report to the Congress the results of its review on obtaining aliens' overseas asset information from the Immigration and Naturalization Service and the future application of this mechanism for reducing

32


aliens' eligibility for SSI benefits. The Department agreed to send this report to the Congress, as we recommended. In September 1980, a Department representative said that a draft report on the alien overseas asset study had been prepared, but that it had not been approved for issuance or release.

The provisions of Public Law 96-265 are expected to reduce the payment of SSI to newly arrived aliens by $12 million in fiscal year 1982 and by $87 million during fiscal years 1981-85. This legislation will not completely resolve the problem discussed in our report-- newly arrived aliens will continue to receive an estimated $32 million during fiscal year 1982 and $118 million during fiscal years 1981-85. However, we believe that if, as proposed, the Department of State adopts the deeming formula of Public Law 96-265 as a guide for determining aged, blind, or disabled visa applicants' eligibility for entry into the United States, such action together with the provisions.of Public Law 96-265 should substantially resolve the problem discussed in our report. In effect, Public Law 96-265 provides that SSI will not be paid to newly arrived aliens whose sponsors are fi- nancially able to provide needed support: and the Department of State's proposed change would, if adopted, generally ex- clude aliens from entry into the United States unless those aliens have sponsors with adequate resources for the aliens' support.

Recommendation to the Secretaries of .--.--I-----------.------ State and Health and Human Services .----------.--

We recommend that the Secretary of State, in cooperation with the Secretary of HHS, develop more stringent income criteria for judging the ability of a sponsor to support a visa applicant.

We also recommend that the Secretary of HHS direct the Commissioner of Social Security to report to the Congress the results of its review on obtaining aliens' overseas asset information from the Immigration and Naturalization Service and the future application of this mechanism for reducing aliens' eligibility for SSI benefits.

33


NEED FOR SYSTEMATIC MEDICAL REVIEW OF DISABLED SSI RECIPIENTS (HRD-78-97, Apr. 18, 1978)

SSA lacks an adequate system for reviewing its SSI disability caseload to insure that only medically eligible persons continue to receive disability payments. When SSI applicants are found to be disabled, a decision is made on whether their impairments are permanent or nonpermanent. For a nonpermanent impairment a reexamination is scheduled when the impairment is expected to improve sufficiently for the person to engage in substantial gainful activity. The estab- lishment of a reexamination date is called a diary.

The SSI disabled population consists of (1) persons who were converted from State disability programs to the SSI program when it became effective in January 1974 and (2) persons who entered the program after that date.

SSA estimated that in 1976, 2.1 million disabled SST. recipients were paid $2.6 billion in benefits. However, only about 70,000 of these recipients were scheduled annually for medical reexamination. By fiscal year 1979 benefits amount- ing to about $3.3 billion were paid to about 2.5 million SSI disabled recipients. Almost 78,000 of these recipients were reexamined during this period. SSA was not routinely monitoring or evaluating recipients who were not covered by a diary to determine whether their impairments improved.

Our report disclosed that a significant number of the disabled SSI recipients selected for review were no longer medically eligible to receive benefit payments. We selected two seperate recipient samples and asked SSA to evaluate the recipients' continued eligibility, which required a review of medical evidence. SSA's evaluation of our first sample of 402 recipients who were converted to the SSI program from State disability programs showed that over half of the cases lacked sufficient evidence to support the disability decisions. SSA found that about one-quarter of the cases with adequate support showed that the recipients were not disabled as defined by appropriate criteria. SSA's evaluation of our second sample of 175 recipients who entered the program after January 1974 disclosed that 10 percent of them were no longer disabled based upon newly obtained medical evidence. Many of the cases in our samples would never have been subjected to medical reexamination under SSA's diary procedures then in existence.


Because the vast majority of SSI disability recipients are not subjected to medical reexaminations, we recommended that SSA establish appropriate mechanisms for systematically reviewing recipients so that those no longer disabled could be removed from the program. In response to our recommendation, SSA is reviewing its medical reexamination procedures to improve their effectiveness. Presently, only recipients with specific impairments are routinely reexamined. SSA is currently revising these procedures to provide physicians who review disability claims the option of scheduling reexaminations where a significant improvement in health can be expected.

SSA is also developing profiles to identify persons most likely to recover from their disability and to select recipients for reexamination. In addition, SSA has initiated several quality control special studies to assess the medical aspects of SSI disability recipients, some of whom are not routinely scheduled for reexamination.

In March 1979 SSA began a nationwide study of the medical eligibility of a national sample of about 13,000 SSI recipients who were converted from State disability programs without Federal review of their disability. This study disclosed that 12.4 percent of these recipients were no longer eligible for benefits. SSA estimates that removal of these recipients from the SSI disability program will result in the following savings.

Year Savings

Annual Cumulative

(millions)

1979 $1.2 $ 1.2 1980 2.5 3.8 1981 2.5 6.2 1982 2.4 8.6 1983 2.2 10.8

However, because the estimated cost to perform this study amounted to about $5 million, SSA will not realize any net savings until 1981. SSA is expanding its nationwide study of converted SSI recipients to include another 100,000 recipients in fiscal year 1981. In addition, SSA is currently performing a special study of such recipients in one State.

35


Public Law 96-265, enacted June 9, 1980, provides for SSA's systematic review of SSI disabled recipients. Sec- tion 311 of this act states that, beginning January 1, 1982, all disability cases, except those where the person is found to be permanently disabled, shall be reviewed at least once every 3 years for the purpose of establishing continued eligibility.

ERRONEOUS SSI PAYMENTS RESULT FROM PROBLEMS IN PROCESSING CHANGES IN RECIPIENTS' CIRCUMSTANCES (HRD-79-4, Feb. 16, 1979)

Recipients of SSI benefits are required to report to SSA changes in income, resources, and other circumstances which could affect the amount of their benefits or continued eligibility for assistance. During our review, SSA was processing over 12 million posteligibility changes annually. We reviewed a sample of these changes and found that 19 percent of the information received by SSA either was lost, was not effectively acted on, or took too long to process.

Our report made several recommendations to strengthen central and district office direction and implement computer controls to ensure that posteligibility changes are properly controlled and processed. In October 1979, HEW advised us that it concurred generally with GAO's recommendations and that SSA had or was taking some actions to implement them. Our followup work, completed in October 1980, showed that SSA had initiated corrective actions as indicated in HEW's comments. However, some of the actions had not been completed.

Our recommendations and SSA's actions to implement them are summarized below.

Recommendation: SSA should establish procedures, goals, and a system for controlling, processing, and monitoring SSI posteligibility changes.

SSA actions: Overall payment error goals and overpayment clearance and collection goals have been established for the SSI program. We were advised that work is underway to establish processing goals for posteligibility changes. Also, several systems changes have been made to improve control and processing of posteligibility changes: other procedural changes are still in process.


Recommendation: SSA should establish pending files for controlling posteligibility changes that are not monitored through the district office workload report system.

SSA actions: SSA officials stated that work was underway to implement uniform filing systems including pending files for posteligibility changes. They stated also that two model case filing systems were being tested by two regional offices to determine what system would be best for SSA. In addition, SSA is in the process of developing a new records management training program for its employees.

Recommendation: SSA should establish procedures to insure that posteligibility information received is processed before it is filed.

SSA action: Several procedural and systems changes have been made, and others are in process to improve controls and processing of posteligibility changes.

Recommendation: SSA should insure that offices retain and dispose of documents in accordance with SSA's records retention and disposition schedules.

SSA action: Most of SSA's record retention schedules, including those relating to posteligibility changes, have been reviewed and revised. SSA officials told us and our review confirmed that their periodic record management surveys would check on how the schedules have been implemented.

Recommendation: SSA should request the National Archives and Records Service to help develop an effective records management program.

SSA action: SSA advised us in October 1979 that it had and would continue to work closely with the National Archives and Records Service to improve its program.

Recommendation: SSA should periodically assess the records management program to determine compliance with the Federal Records Act.

37

.

1 ‘_’

‘. ‘,. . . : <’ ;

, . “ . :

APPENDIX II APPENDIX II'

SSA action: We were advised that SSA has a process for periodically assessing its field offices' records management programs and will continue to make cyclical onsite assessments. SSA has issued six reports, the last of which was issued in October 1979, based on its assessment of the records management program of 18 district offices in six regions.

Recommendation: SSA should establish controls in the computer system to assure field offices that all posteligibility changes transmitted by them are either posted to the record or rejected.

SSA action: We were advised that work is underway to develop computer specifications for better control of posteligibility changes transmitted by field offices. Implementation of improved computer controls should be completed late in 1981.

Recommendation: SSA should establish controls over rejects so that the system can notify field offices when information in reject messages has not been corrected.

SSA actions: SSA officials told us that some system changes had been made to improve controls and processing of posteligibility changes including rejects. We were advised that a project was underway specifically to develop system specifications for better control and processing of rejects. If approved, the system changes should occur early in 1982.

Recommendation: SSA should evaluate the alert system to insure its effectiveness.

SSA action: According to SSA officials, in June 1980, SSA initiated a study of uncontrolled posteligibility changes to determine the magnitude of the problem and the cost savings that would result from better controls. We were advised that preliminary study results show that (1) rejects and alerts appeared to be causing most of the problems, (2) p recessing problems continue to exist, (3) better management and controls are needed, and (4) better controls would result in cost savings, We were advised also that systems specifications were being drafted for better control and processing of alerts and if accepted should be implemented sometime by mid-1981.

38


Recommendation: SSA should reemphasize to field offices the need to process rejects and alerts.

SSA actions: SSA advised us in August 1980 that it was developing system specifications for better control and processing of rejects and alerts which would also emphasize the need to process these items in a timely manner. If approved the changes for alerts and rejects are expected to be implemented early in 1981 and 1982, respectively.

Recommendation: SSA should periodically monitor the field otfices to insure that rejects and alerts are promptly and effectively processed.

SSA actions: We were advised that checks and balances have been included in proposed system changes to help the central office monitor rejects and alerts. In addition, the agency's management information system is being revised to provide more information on the handling of these two events.

Recommendation to Secretary of Health and Human Services

We recommend that the Secretary direct the Commissioner of SSA to complete actions on those recommendations not yet fully implemented.

SSA SHOULD IMPROVE ITS COLLECTION OF OVERPAYMENTS TO SSI RECIPIENTS (HRD-79-21. Jan. 16. 19191 --

We reported problems that SSA had in resolving SSI overpayments. These problems included:

--A massive backlog of unresolved SSI (about $462 million as of September

over ayments 1978 P .

--An inadequate method of notifying overpaid recipients.

--Inconsistent application of overpayment resolution policies and procedures.

We made the following recommendations to HEW to correct these problems and improve collection of overpayments.

39


--Establish standards for timely processing of SSI overpayments.

--Develop a quality control mechanism to assure performance in accordance with SSA's overpayment policies and procedures.

--Develop improved instructions and additional training in overpayment resolution for claims representatives to assure that all overpayment cases are treated more uniformly.

--Develop an automated notice to inform recipients when they have been overpaid, the cause of the overpayment, proposed agency actions, and the recipient's appeal rights.

--Develop, through use of the quality control mechanism, more useful and less subjective criteria to determine whether an overpaid recipient was with or without fault in causing the overpayment.

We also recommended that, once these recommendations are implemented and there has been a measurable improvement in SSA's overpayment collection process, the Secretary seek legislation to authorize offsetting SSI overpayments against title II and other Federal benefit-paying programs.

Several efforts have been initiated to resolve the problems we identified. SSA has developed proposed goals for processing SSI overpayments. However, these goals are not yet finalized.

SSA made a pilot study to test a quality control mechanism to assure performance with SSA's policies and procedures and decided it will not commit resources to a continuing national quality control review of SSI overpayments.

SSA cited several factors which influenced its decision. One reason given was that quality control reviewers would be superimposing their judgment over that of the decisionmakers. In addition, SSA believed that its limited staff resources would be more productive in the analysis and review of initial and redetermination cases. Finally, SSA believed that the management information presently available could ade- quately pinpoint problems in the overpayment process. We still believe that a quality control mechanism should be implemented, as we recommended. We believe that the type of

40

. . “ . ’ 2’.

/’ ( : :

..’

; : , , . . ._ . .

>, I / ,


information this mechanism could generate is important to identifying problems on a continuing basis so that management will be able to take timely corrective action.

SSA has also revised its instructions and provided additional training to field staffs. For instance, SSA's claims manual has been revised to give more examples of with or without fault determinations, and sections of the manual were rewritten in clearer language. SSA has also developed training programs on SSI overpayment resolution activities including waiver decisions, compromise settlements and refund controls. These programs (which include videotapes) have been provided to all program service centers, district offices, and branch offices.

Action on our fourth recommendation is not complete; SSA is still developing specifications for an automated notice to overpaid SSI recipients. An SSA representative estimated that it will be about January 1982 before the notice can be tested.

Although SSA has revised its claims manual to give more examples of with or without fault determinations, SSA has not developed a quality control mechanism to provide continuing information highlighting field offices' problems and inconsistencies in interpreting the with or without fault provision. We believe this information is needed to enable SSA to revise its claims manual criteria and examples to minimize those problems and inconsistencies.

SSA has made some improvement in its actual collection efforts. For instance, SSA had collected about 20 percent of all SSI overpayments as of September 1978 and about 26 percent of all SSI overpayments as of June 1980. In addition, there has been a decrease in the amount of unresolved overpayments. However, SSA still waives about 30 percent of all SSI overpayments, and as of June 1980, about $346 million in SSI overpayments were unresolved.

Although SSA has considered proposing legislation to allow SSI overpayments to be offset against title II benefits, the Department has not submitted this proposed legislation to the Congress. While SSA has made some improvement in collecting SSI overpayments and partially implemented our recommendations, we believe that our recommendations should be fully implemented and further improvements achieved in recovering SSI overpayments before such legislation is proposed to the Congress.

41

I , , : ‘.

. , “J * I . , .

. . , . I

; . - .

. ; - 1 :’

‘i


The actions taken by the Department have apparently had beneficial effects on the collection of SSI overpayments. However, it is not possible to quantify the savings attributable to these actions or the potential savings which would result from full implementation of our recommendations.


We recommend that the Secretary direct the Commissioner of SSA to adopt a stronger and more active management role in recovering SSI overpayments by:

---Establishing standards for timely processing of SSI overpayments.

--Developing a quality control mechanism to assure performance in accordance with SSA's overpayment policies and procedures.

--Developing an automated notice to inform recipients when they have been overpaid, the cause of the overpayment, proposed agency actions, and the recipient's appeal rights.

--Developing-- through use of the quality control mechanism-- more useful and less subjective criteria to determine whether an overpaid recipient was with or without fault in causing the overpayments.

When these recommendations are fully implemented and there has been further improvement in the SSI overpayment collection process, the Secretary cjhould seek legislation to authorize offsetting overpayments against title II and other Federal benefit-paying programs.

IMPROVEMENTS NEEDED TO INSURE ACCURACY OF SSI RETROACTIVE PAYMENTS (HRD-79-26, Dec. 11, 1978)

SSA is generally required to pay eligible SSI applicants from the first of the month in which they apply. If approval takes several months, the initial payment will cover the entire retroactive period from the first day of the month of application.

42

/‘, _ ‘. ,“‘l

, I_

. , ‘.

J


We reported that during fiscal year 1977 SSA overpaid recipients of retroactive SSI benefits about $75 million and underpaid them about $6 million. Most of the payment errors occurred because SSA was not obtaining accurate information affecting the claimant's eligibility or amount of benefit. This happened because SSA was unaware of changes in the claimant's circumstances which occurred between the date of the application and the date of payment.

In August 1977, because of concern about inaccurate retroactive payments, SSA initiated a special prepayment review of all initial and posteligibility payments of $5,000 or more. In fiscal year 1978, this process identified errors in about 40 percent of the payments reviewed and prevented overpayments of about $1.4 million and underpayments of about $65,000.

We reported that SSA had developed priority redetermination procedures for all initial claims cases where appeals to higher levels resulted in favorable decisions for applicants. However, when first implemented, the redeterminations were usually done after payment had been made. In September 1978, SSA implemented a new policy whereby most of these cases would be subject to priority redeterminations before payments are made.

,

We recommended that- SSA evaluate the cost effectiveness of expanding its prepayment reviews to include more retroactive payments and priority redeterminations. We recommended also that analysis of retroactive payment errors be included in the Quality Assurance Office's review and that goals be established to reduce retroactive payment errors.

HEW advised us in August 1979 that actions had been or were being taken to implement our recommendations. HEW said that:

--Beginning in February 1979, all retroactive payments of $3,000 or more had been subjected to prepayment reviews.

--Beginning in July 1979, retroactive payments in the $2,000 to $3,000 range would be reviewed.

--Revised instructions would be issued to better control payments in the $1,000 to $2,000 range.

43


--Goals would be established for reducing retroactive payment errors as it evaluates its prepayment review efforts.

--It would explore ways in which detailed analysis of retroactive payment errors could be made part of ongoing quality assurance reviews.

--It would immediately start an experiment to test the cost effectiveness of requiring priority redeterminations before any payments of $2,000 or more are made.

Information obtained from SSA in August 1980 shows that, while the Quality Assurance Office's review has been revised to include data on retroactive payment errors, goals have not been established to reduce such errors. Also, SSA has evaluated the cost effectiveness of expanding its prepayment reviews. The cost study indicated that expanding these reviews would be cost effective. However, we were advised by SSA officials in August 1980 that, although the agency had initiated a central office prepayment review of all retroactive payments of $3,000 and over, the entire process was involved in a major reorganization. Information obtained from SSA shows that the process was stymied by disagreement on the organizational location and staffing of different segments of the prepayment review function. As a result, there was less prepayment review activity in January 1980 than there had been when HEW prepared its response to GAO's report.

SSA officials told us that from February 1979 through about March 1980 the level of retroactive prepayment review fluctuated between $3,000 and $5,000 depending on the avail- ability of staff. They said that since about March 1980 these reviews have been at the $5,000 and above level except for forced payments and automated one-time payments of $3,000 or more. In addition, the New York and San Francisco regions had established special review units to review payments involving administrative law judges' decisions and other selected cases.

SSA advised HEW's Acting Inspector General in April 1980 that the corrective actions on our recommendations "are not on track," but that the functional-organizational issues which stymied progress had been resolved. In September 1980 SSA officials stated that a central office prepayment review group had been established to centrally review all retroactive payments of $3,000 or more, and the group was expected to reach full staffing in mid-1981.

44


SSA also advised the Acting Inspector General that, after its $3,000 level reviews are underway, an analysis will be made of payments in the $2,000 to $3,000 range to determine the cost effectiveness of a centrally directed prepayment review of all retroactive payments of $2,000 or more. In addition, SSA stated that further actions on our other recommendations would depend on the accumulation and analysis of the operational data obtained from ongoing prepayment reviews.

In May 1980, SSA estimated that prepayment reviews of all retroactive payments of $5,000 or more already being made would result in annual reductions in overpayments of about $1.3 million. Using the data provided by SSA, we estimated that these prepayment reviews would cost about $146,000, leaving a net annual savings of about $1.2 million. The data provided by SSA also show that lowering the prepayment review level from $5,000 to include all payments of $3,000 or more would have resulted in additional annual savings of about $1 million, and a reduction to the $2,000 level would have resulted in an additional net savings of about $1 million. SSA's cost estimates did not include data on the amount of underpayments that would be identified as a result of these prepayment reviews.


We recommend that the Secretary direct the Commissioner of SSA to:

-pReview all retroactive SSI payments of $2,000 or more.

--Establish goals for reducing retroactive payment errors.

SSI OVERPAYMENTS DUE TO SSA'S DETERMINING ELIGIBILITY AND BENEFITS ON A PROSPECTIVE QUARTERLY BASIS (~~~-78-114, May 26, 1978)

Currently, continuing eligibility for and the amount of SSI benefits are determined on a prospective quarterly basis. SSI benefits are based on the non-SSI income SSI recipients expect to receive over a projected 3-month period. Once computed, these payments are disbursed in equal monthly installments. The prospective quarterly

45


computations were established to minimize changes in the monthly benefit payments caused by variations in recipients'

\ income.

We reported that substantial overpayments to SSI recipients have occurred because eligibility and benefit amounts are determined on the prospective quarterly basis. Requiring recipients to estimate future changes caused by such things as death, marriage, separation, or divorce, entering or leaving an institution, and earned or unearned income has resulted in inaccurate monthly benefit payments and administrative difficulties. Further, recipients find it difficult to understand why they were overpaid when they reported the receipt of unexpected income during a quarter. We reported that in 1976 alone SSA estimated that at least $39 million of SSI overpayments were caused by this accounting method. Such overpayments are expected to increase to about $60 million in 1982.

We recommended that the Social Security Act be amended to require that a retrospective monthly accounting period be used as the basis for determining SSI benefit eligibility and payment amount. Further, we recommended a l-month lag between the month used for eligibility determinations and benefit amount calculations, and the month payments are made to recipients. We stated these proposed changes should substantially reduce erroneous payments caused by the present prospective quarterly accounting period and reduce SSA's burden of developing and processing overpayments. However, our proposed accounting methodology could delay payments to initial applicants for up to 2 months, depending on the date the applicant filed for SSI benefits and the processing time required by SSA. To avoid this delay, we recommended that such applicants should be allowed to file an application for such benefits 1 month prior to their date of eligibility, and the benefit amounts should be based on income and benefit status in the month prior to application.

On June 5, 1978, HEW submitted to the Congress proposed Social Security Act amendments cited as the "Social Welfare Reform Amendments of 1979." Section 232 of the proposed amendments provided that SSI eligibility and benefit amounts would be determined on a retrospective monthly basis rather than a prospective quarterly basis. This proposed amendment agrees with that part of our recommendation concerning a retrospective monthly accounting period, but is silent regarding the l-month lag period we recommended. An SSA official stated that it was not feasible to set an exact

46


l-month time period between the month used for eligibility determinations and benefit calculations and the month payments are made to recipients. He said it might take more or less than 1 month to determine eligibility and benefits and process payments, and by not including a l-month lag period, the agency intended to make the time interval as short as possible. SSA estimated cost savings to the Fed- eral Government in fiscal year 1982 of $60 million if this proposed amendment had been approved.

On November 7, 1979, the House of Representatives passed a modified version of HEW's proposed welfare reform legislation (H.R. 4904). Section 232 provides for determining SSI eligibility and benefit amounts on a monthly basis rather than a quarterly basis. However, this section does not incorporate the retrospective basis for determining SSI eligibility and benefit amounts as we recommended, and HHS estimates that no savings would accrue to the Government from enactment of section 232 of the House-passed version of H.R. 4904.


We recommend that the Congress amend the Social Security Act to provide for determining SSI benefit eligibility and payment amounts on a monthly retrospective basis rather than the quarterly prospective basis.

SSI OVERPAYMENTS TO MEDICAID NURSING HOME RESIDENTS CAN BE itEDUCED m-7-131, Aug. 23, 1977)

When an SSI recipient enters a nursing home for an expected stay of a full calendar month or longer, SSI payments should be reduced to not more than $25 for each month of residence, because the cost of nursing home care is paid through the Medicaid program. The $25 is for personal and incidental expenses which are not covered by Medicaid.

We reported that SSI overpayments were being made because SSA often did not know that recipients had been admitted to nursing homes. In California and Florida such overpayments amounted to $7.6 million during 1975. The majority of these overpayments could have been prevented through timely reporting of nursing home admissions. SSA's reliance on recipients to report admissions has not been effective, Our study showed

47


that SSI recipients, or the persons authorized to accept payments on their behalf, reported only 3 percent of nursing home admissions.

We recommended that SSA (1) require its district offices to provide forms to nursing homes for reporting admissions and (2) actively work with the nursing homes to obtain timely reports. HEW concurred with this recommendation and revised a preaddressed and postage paid form to provide for reporting the date an SSI recipient entered or left an institution such as a nursing home. Copies of the form were provided nursing homes, and SSA required its district offices to encourage nursing homes to promptly report admissions of SSI recipients by using the preaddressed form.

We also recommended that the Health Care Financing Admin- istration have States establish procedures requiring nursing homes participating in the Medicaid program to promptly report admissions of SSI recipients to district offices. HEW said that it did not have legal authority to require nursing homes to report directly to SSA. Instead, HEW said it would implement an alternative plan to require nursing homes participating in the Medicaid program to report all admissions of SSI recipients to State Medicaid agencies which would be required to report this information monthly to SSA regional offices. However, an SSA representative informed us that this alternative plan was never implemented, and information supplied by SSA regional offices shows there is no uniform procedure in effect for assuring that SSA is notified when SSI recipients enter nursing homes.


We recommend that the Secretary (1) direct the Commis- sioner of Social Security to make a nationwide review to identify nursing homes failing to complete and return the reporting form when SSI recipients are admitted and (2) direct the Administrator of the Health Care Financing Administration to have States establish procedures requiring those nursing homes to promptly report the admission of SSI recipients to the State Medicaid agencies, and require those agencies to promptly remit such information to SSA.

49

.I’ ., ‘Y

, I : : , :

. . : , ,’ , : 1

‘.. , . . . , . : _. ‘. , : . .

, ‘., : .

j

' APPENDIX II APPENDIX II

NEED TO ASSESS PENALTIES AGAINST SSI RECIPIENTS WHO FAIL TO REPORT CHANGES AFFECTING ELIGIBILITY AND BENEFIT PAYMENTS (HRD-78-118, May 22,1978)

Substantial overpayments have been made to SSI program recipients who have not reported changes in their income, resources, or other circumstances to SSA. The integrity of the SSI program depends, to a large degree, on accurate and timely reporting of these changes which can affect not only the amount of benefit payment but also continued eligibility for assistance.

The Congress, in establishing the SSI program, gave SSA authority under the Social Security Act (42 U.S.C. 1383 (e) (2)) to assess penalties against recipients who do not fulfill their reporting responsibilities. According to Federal regulations (20 C.F.R. 416), changes should be reported as soon as they occur. Failure to report changes within 30 days after the quarter in which they occur may cause penalties to be imposed unless there is good cause for not reporting the changes.

SSA had taken steps to increase SSI recipients' aware- ness of their reporting responsibilities and the possible imposition of penalties. Periodically, the agency mailed informational pamphlets to recipients providing them with information on the SSI program and their reporting responsibilities. SSA instructed its field offices to inform recipients about timely reporting requirements and penalties that may be imposed for untimely and nonreporting of changes. In addition, the claims manual for the SSI program instructed field offices on the steps to be taken in imposing and collecting penalties for failure to report or late reporting. However, we found that the computer system which processes SSI benefit payments had never been programed to process penalty actions that would be submitted by the field offices.

Because the accuracy of SSI payments to recipients is highly dependent upon their timely reporting of changes, we recommended that SSA be directed to implement the penalty provision, including necessary programing of the computer system, to process penalties for untimely reporting and nonreporting of changes in recipient circumstances. SSA implemented our recommendation in November 1978. According to agency records for the period November 1978 through July 1980, penalties totaling $138,450 have been assessed, of which $118,359 have been collected.

49

APPENDIX II APPENDIX I,1

We also recommended that recipients be notified when actual implementation of penalty deductions will begin and the amounts involved for initial and repeat violators. In December 1978, SSA notified SSI recipients of its plans to assess penalties for failure to meet reporting requirements and the penalty amounts provided by law. Implementation of our recommendations should improve recipient compliance with reporting requirements and reduce SSI overpayments. SSA officials told us that the exact amount of savings cannot be determined.

APPENDIX III APPENDIX III

ACTIONS TAKEN ON RECOMMENDATIONS CONCERNING THE

AID TO FAMILIES WITH DEPENDENT CHILDREN AND

EMERGENCY ASSISTANCE PROGRAMS

Aid to Families with Dependent Children is one of the largest federally added public assistance programs. Admin- istered by the States in cooperation with HHS, the program provides financial assistance to needy children and their parents or relatives to encourage the care of dependent children in their homes.

Each State must submit a comprehensive plan to HHS de- scribing the nature and scope of its AFDC program and its promise to administer the plan according to Federal statutes and regulations. Federal and State payments for AFDC during fiscal year 1981 are estimated to total about $12.8 billion, of which the Federal share will be about $6.9 billion or 54 percent. The Federal share varies among States and ranged from 50 to 78 percent in 1980.

The Emergency Assistance program was established to provide financial assistance and social services to needy families with children under 21 to meet emergency needs. To receive Federal funds, States must include emergency assistance in the plans for their AFDC programs. The Federal Gov- ernment pays for 50 percent of the assistance provided and the related administrative costs. During fiscal year 1981, total emergency assistance payments are expected to be about $92.5 million; the Federal share will be about $46.2 million.

AFDC OVERPAYMENTS CAUSED BY DELAYS IN STOPPING PAYMENTS TO INELIGIBLE RECIPIENTS (HRD-78-87. Mar. 22, 1978)

On October 21, 1976, we reported to the Secretary of HEW that Ohio and New York City were making erroneous payments to individuals who were no longer eligible for AFDC benetits. Ohio misspent an estimated $5 million in 1 year, and the New York State Comptroller estimated that New York City misspent about $9 million annually because payments could not be stopped in a timely manner.

In a followup report to the Secretary on March 22, 1978, we said that HEW determined that Ohio had taken appropriate corrective action. Ohio had implemented a system for county welfare departments to report eligibility changes through

51


regional computer terminals. We estimated the resulting annual savings was $5 million ($2.7 million in Federal funds and $2.3 million in State funds).

We also reported in 1978 that HEW had not implemented the following recommendations in our 1976 report.

--Examine the termination process in New York City to determine if the problems have been corrected.

--Determine whether other States could realize savings through faster AFDC terminations.

HEW did not help New York City correct the problems identified in our 1976 report. However, New York City took action to reduce its delays in terminating AFDC payments and estimated annual savings to be $8.9 million, one-half of which is Federal funds.

The actions taken by New York City to improve its termination of AFDC payments included changes in internal mail handling, closer monitoring of the time needed to stop the payments, and changes in the procedures required by New York State for notifyng recipients of impending terminations. Both New York State and New York City officials stated that New York City.had achieved its greatest savings from improved computer capabilities --the changes in AFDC eligibility could be entered locally on remote computer terminals instead of being hand carried to a central location.

HEW indicated in January 1977 that it planned to review the AFDC payment termination process in all States. In October 1980, HHS officials stated that preliminary efforts in reviewing the termination process in five States indicated that none were experiencing the same problems. A survey conducted in 20 additional States was terminated when the survey report by a contractor was found unacceptable. According to HHS, the study will not be finalized, and no further effort in this area is planned.


We recommend that the Secretary determine whether other States have similar problems in stopping AFDC payments to ineligibles, and if so, help those States to correct them.

52


REQUIREMENTS AND PRACTICES FOR REFUNDING THE FEDERAL SHARE OF UNCASHED AFDC CHECKS (HRD-79-68, Apr. 5, 1979)

We reviewed HEW requirements and the practices of States for refunding or crediting the Federal Government's portion of checks that were issued to AFDC recipients but never cashed. Our review was made in Massachusetts and New York, and we obtained information from nine other States. Although New York credited the Federal Government for its share of uncashed checks in most instances, about $4 million had not been credited to the Federal Government by Massachusetts for AFDC checks issued but not cashed from fiscal year 1968 through fiscal year 1977. As a result of our review, Massachusetts made this credit to the Federal Government. In addition, HEW had previously found that Illinois and Puerto Rico had not credited the Federal Government for its portion of uncashed AFDC checks.

HEW had not provided criteria to govern States' reim- bursement of the Federal portion of uncashed AFDC checks. The only guidelines HEW had provided were instructions for making general credits, such as credits for recoveries of program funds overpaid to recipients. Some State and local agencies have established their own criteria for such credits, but there is no assurance that appropriate credits have or will be made.

We recommended that the Secretary of HEW direct the Commissioner of Social Security to establish (1) uniform requirements for States to credit the Federal Government for its portion of uncashed AFDC checks and (2) a mechanism for insuring that these credits are timely and accurate.

In responding to our report on April 23, 1979, HEW agreed to issue a Federal regulation establishing uniform requirements for States to credit the Federal Government for its portion of uncashed checks that have been previously paid with Federal funding. In addition, HEW stated that it would insure that such credits were timely and accurate by establishing a mechanism that, when implemented, would require States to account quarterly for all checks not cashed within a specified period.

APPENDIX III APPETDIX III

As an interim measure, on June 15, 1979, HEW instructed its regional offices concerning the necessity for them to insure that States identify and make timely Federal AFDC credits. Each region was requested to perform a survey of procedures followed by each State within their region. Most regions completed their survey in August 1979. Although the results varied by State, the survey confirmed the need for a uniform policy in crediting Federal programs. In May 1980, the Department's Audit Agency in the Office of Inspector General reported that their reviews also indicated a problem in the timely return by States of the Federal portion of uncashed checks and concurred with our recommendations. The Audit Agency stated that the lack of a uniform policy has a substantial monetary impact on Federal programs.

As of October 1980, HISS had not issued the planned regulations establishing the uniform requirements or the mechanism for insuring the accuracy and timeliness of credits. HHS officials stated that the regulations are being drafted and States will have an opportunity to comment on them before they are issued. HHS does not plan to establish the mechanism for insuring timely and accurate credits until after issuance of these regulations.

In our report we also recommended that HEW identify and recover the total amount of Federal funds in uncashed AFDC checks that have not been refunded to the Federal Govern- ment. In implementing this recommendation, HEW requested its Audit Agency to review State systems and procedures for refunding the Federal portion of uncashed AFDC checks. Since October 1, 1979, the Audit Agency has completed reviews in 9 States and has ongoing reviews in 19 additional States. Its reviews have shown that State laws vary considerably with respect to their policies for cancelling checks and crediting Federal programs, and in some cases, there was no provision for returning Federal funds relating to uncashed checks.

In seven of the nine State reviews completed, the Audit Agency identified $1.6 million as the Federal share of uncashed AFDC checks that should be returned to the Federal Government. In most cases, States involved have agreed to make the appropriate credit.

In four of its ongoing reviews, the Audit Agency has identified about $880,000 that it will attempt to recover. No information was available concerning the expected recoveries from the other 15 State reviews in process. HI-IS plans

54


to initiate reviews in 14 additional States during fiscal year 1981, and the remaining States will be reviewed after fiscal year 1981.

Recommendations to the Secretary ----- of Health and Human Services __----.----_L__-.__-

We recommend that the Secretary direct the Commissioner of Social Security to issue regulations establishing uniform requirements for States to return the Federal portion of AFDC checks not cashed and establish a mechanism for insuring that these credits are timely and accurate.

NEED FOR IMPROVED OVERPAYMENT RECOVERY ---.-_I POLICIES IN THE AFDC PROGRAM (HRD-78-117, May 25, 1978)

SSA estimated that during 1976 over $850 million, about one-half of which is Federal funds, was erroneously paid to AFDC recipients. Erroneous payments were made for several reasons, including fraud in applying for AFDC and the willful withholding of information by recipients about their need for continued assistance. The growing magnitude of erroneous payments is illustrated by Federal quality control samples, which show that more than $1 billion was overpaid AFDC recipients from April 1978 through March 1979.

Recovering these overpayments was generally left to the States' discretion. Some States required recipients to fully repay overpayments. Others either waived the amount overpaid, sought voluntary repayment, or attempted recovery only if fraud was involved.

PrOViSiOnS of the Social Security Act governing the AFDC program do not address the issue of recovery of AFDC overpayments. However, the Congress has given HHS broad authority to establish regulations to enable States to properly and efficiently administer and operate their AFDC programs. Thus, HHS has the authority to require States to establish uniform and comprehensive provisions for recovering all types of AFDC overpayments and for maintaining accountability and control of overpayments.

We recommended that HEW revise its regulations to establish uniform and comprehensive overpayment recovery policies in the AFDC program for all types of overpayments. We sug-

9 ested that the regulations include a requirement that States 1) maintain information on the total number and amount of

55

APPENDIX XII APPENDIX III. ,

As an interim measure, on June 15, 1979, HEW instructed its regional offices concerning the necessity for them to insure that States identify and make timely Federal AFDC credits. Each region was requested to perform a survey of procedures followed by each State within their region. Most regions completed their survey in August 1979. Although the results varied by State, the survey confirmed the need for a uniform policy in crediting Federal programs. In May 1980, the Department's Audit Agency in the Office of Inspector General reported that their reviews also indicated a problem in the timely return by States of the Federal portion of uncashed checks and concurred with our recommendations. The Audit Agency stated that the lack of a uniform policy has a substantial monetary impact on Federal programs.

As of October 1980, HHS had not issued the planned regulations establishing the uniform requirements or the mechanism for insuring the accuracy and timeliness of credits. HHS officials stated that the regulations are being drafted and States will have an opportunity to comment on them before they are issued. HHS does not plan to establish the mechanism for insuring timely and accurate credits until after issuance of these regulations.

In our report we also recommended that HEW identify and recover the total amount of Federal funds in uncashed AFDC checks that have not been refunded to the Federal Govern- ment. In implementing this recommendation, HEW requested its Audit Agency to review State systems and procedures for refunding the Federal portion of uncashed AFDC checks. Since October 1, 1979, the Audit Agency has completed reviews in 9 States and has ongoing reviews in 19 additional States. Its reviews have shown that State laws vary considerably with respect to their policies for cancelling checks and crediting Federal programs, and in some cases, there was no provision for returning Federal funds relating to uncashed checks.

In seven of the nine State reviews completed, the Audit Agency identified $1.6 million as the Federal share of uncashed AFDC checks that should be returned to the Federal Government. In most cases, States involved have agreed to make the appropriate credit.

In four of its ongoing reviews, the Audit Agency has identified about $880,000 that it will attempt to recover. No information was available concerning the expected recoveries from the other 15 State reviews in process. HHS plans

54


$33.6 million in Massachusetts and a minimum of about $6 million in New York. In addition, about $1.4 million of the $33.6 million and an undeterminable portion of the $6 million are overpayments which may not be recouped.

Massachusetts provided AFDC recipients with a portion Of their assistance payment in advance payments four times a year. The advances are recouped in full through reductions in subsequent regular monthly payments. The advance payment policy appears inconsistent with the Code of Federal Regula- tions (45 C.F.R. 233.10 (b)(3)) which provides for Federal participation only if the recipient was eligible on the date aid was paid. Federal participation should therefore be claimed only for that portion of a quarterly advance payment which applies to those months in the quarter the recipient was eligible, rather than the entire quarter. However, Massachusetts has claimed Federal participation for the total amount of the advance payments. Furthermore, Massachusetts does not require recipients to repay an advance payment, or a proportionate amount, if they become ineligible at any time during the quarter.

We recommended that HEW revoke its approval of Massa- chusetts' quarterly advance payment policy and limit Federal participation to payments for those months in each quarter that each recipient was eligible.

HHS responded to our report on May 12, 1980, by agreeing with the recommendation to discontinue the quarterly advance payment practice in Massachusetts. On May 28, 1980, Massa- chusetts took action effective July 1980 to discontinue these quarterly payments: this resulted in recurring savings of at least $1.5 million annually ($778,000 Federal share).

New York authorizes advance payment of AFDC funds for eligible recipients who face eviction or utility shutoffs for overdue bills. These advance payments are in addition to the regular monthly grants and are, in effect, interest- free loans which must be paid from future monthly grants.

New York does not limit the size, number, or total amount of advances a recipient can obtain and have outstanding and did not know the statewide total of these advance payments. According to New York City program officials, about $6 million in advance payments were received by the city's AFDC recipients during 1978. Although these advance payments are subject

57


overpayments involved and their disposition and (2) establish a mechanism for assessing the effectiveness of their overpayment recovery efforts. In October 1978, SSA officials said a regulation would be issued establishing an overpayment recovery policy. However, 2 years later, the regulation has not been issued. SSA officials said that the regulation is being prepared and should be published for comment by January 1981.

We also recommended that SSA assist States in establishing an appropriate mechanism for monitoring and evaluating the adequacy and effectiveness of their recovery efforts and periodically review States' compliance with the requirements established in the regulations. HEW, in response to this recommendation, stated that the regulatory change process takes time to complete. Therefore, SSA established an interim national goal to improve State performance in recovery of overpayments: however, SSA does not know whether States have improved their performance.


We recommend that the Secretary:

-*Revise the regulations to establish uniform and comprehensive overpayment recovery policies in the AFDC program for all types of overpayments, including requirements for States to (1) maintain information on the total number and amount of overpayments involved and their disposition and (2) establish a mechanism for assessing the effectiveness of their overpayments recovery efforts.

-TDirect that SSA help States establish an appropriate mechanism for monitoring and evaluating the adequacy and effectiveness of their recovery efforts and periodically review States' compliance with the requirements established in the regulations.

STATE ADVANCE PAYMENTS TO AFDC RECIPIENTS ARE INCONSISTENT WITH FEDERAL REGULATIONS -(HRD-8Tzmeb. 7, 1980)

Massachusetts and New York have been making payments to AFDC recipients in addition to their regular monthly benefits-- a practice which we believe to be inconsistent with Federal regulations. During 1978, these payments amounted to about

56



We recommend that the Secretary disallow any claims for Federal participation in the advance AFDC payments program of New York and Oregon and initiate appropriate efforts to recover the Federal share of any of these advance payments.

SHOULD EMERGENCY ASSISTANCE FOR NEEDY FAMILIES BE CONTINUED? IF SO, PROGRAM IMPROVEMENTS ARE NEEDED -m-65, Apr.5,iS)

The Emergency Assistance program was authorized by the Congress in 1967 to provide Federal funds to States for temporary emergency assistance to needy families with children. States may either provide cash or arrange for the provision of such items as food, clothing, rent, utilities, or medical care. The Federal Government pays 50 percent of the assistance provided and the related administrative costs.

Our review included four of the largest users of emergency assistance funds--New York, Ohio, Maryland, and the District of Columbia. Together, these four users' total program expenditures during fiscal year 1976 amounted to $39.3 million or 59 percent of all program expenditures.

We found that HEW allowed States wide latitude in developing their Emergency Assistance programs and had not developed uniform guidelines for approving and monitoring State plans. As a result, HEW experienced some operational problems and States used emergency assistance funds for questionable purposes. Also, as of September 1977, 30 States were not participating in the program.

In September 1977, HEW submitted proposed legislation to the Congress which it believed would help resolve the problems of legal definitions and interpretations. The proposed legislation provided for participating States to define the scope of their Emergency Assistance programs. A State would be able to (1) prescribe the categories of needy families with children that could participate in the current program and (2) specify the types of emergencies under which a family could be eligible for assistance. It would also establish a separate Emergency Assistance program for needy families--with or without children-- and for individuals in the case of a pres- identially declared natural disaster or other occurrence of regional or national significance beyond the States' control.

59


to repayment from future monthly grants, neither city nor State officials could tell us how much had been repaid or the outstanding amounts.

The Code of Federal Regulations (45 C.F.R. 233.20(b)(l)) provides for Federal participation in such a program only if the recipient's income plus the monthly AFDC payment does not exceed the need standard --the money amount specified in the State plan to be used in determining the need of applicants and the amount of assistance payments. We concluded that New York's advance payment practice was inconsistent with these regulations because the additional moneys are (1) more than the need standard in the approved State plan, (2) for expenses covered by prior months' grants, and (3) based on the assumption that a recipient will be eligible for future AFDC payments.

We recommended that the Secretary of HEW disallow any claims for Federal participation in New York's advance AFDC payments. We also recommended that the Secretary initiate appropriate efforts to recover the Federal share of any of these advance payments.

HHS has taken no action to stop New York from making advance AFDC payments. In October 1980, HHS informed us that meetings were being held with New York State officials to discuss the State's advance AFDC payment policy and to determine whether the State could possibly modify its program so that it could continue to claim Federal sharing for the costs of providing needed assistance to AFDC recipients facing eviction or utility shutoffs.

In view of our findings in Massachusetts and New York, we also recommended that (1) the Secretary of HEW require SSA to review all State AFDC plans and regulations to see whether their payment policies are consistent with the Code of Federal Regulations and (2) establish a mechanism to insure that changes are made to those State plans with payment policies not consistent with the code.

According to HHS, SSA's review of all State AFDC plans showed that Oregon was the only additional State with an advance AFDC payment policy. HHS plans to follow up with regional officials to determine the appropriate action to be taken with respect to this State's advance payments.

58

I APPENDIX III APPENDIX III

HEW issued guidelines in December 1978 to States which dealt with the approval of State plans, and it has drafted instructions which clarify the degree of latitude States have in defining who is eligible for coverage under the Emergency Assistance program. These guidelines and draft instructions were based on HEW's interpretation of a June 6, 1978, Supreme Court ruling (Quern v. Mandley, 436 U.S. 725) and other court --- decisions which appeared to support HEW's policy of allowing States to define the types of emergencies to be covered under emergency assistance and of providing States with latitude in determining the needy families with children who can participate. HEW has not developed uniform guidelines concerning the uses of funds in the program and how State programs should be monitored to assure uniformity in the use of funds.

The legislation proposed by HEW in September 1977 was not enacted. H.R. 4904, introduced July 23, 1979, which includes a definition of the circumstances under which emergency assistance may be provided, passed the House with amendments, but as of October 1980 had not been acted on by the Senate.

Recommendations to the Congress ---

The Congress should consider whether the Emergency Assistance program should continue. If the Congress determines that the program should continue, it should review the positions of HHS and the courts, including the U.S. Supreme Court, concerning eligibility and the type and extent of emergencies covered. It should then, if necessary, amend the legislation to clearly indicate congressional intent.

LEGISLATION NEEDED TO IMPROVE PROGRAM FOR -.- REDUCING ERRONEOUS WELFARE PAYMENTS -m-164,-iig. 1, 1977)

BETTER MANAGEMENT INFORMATION CAN BE -- OBTAINED FROM THE QUALITY CONTROL SYSTEM -------------I -- USED IN THE AFDC PROGRAM THRD-80-80, July 18, 1980)

Erroneous payments to AFDC program recipients have been a major concern. Two of our reports dealt with efforts to reduce erroneous AFDC payments. During our first review, nearly $500 million in Federal funds was being misspent annually.

61


We stated that allowing States to define the scope of their programs would not resolve the problem of questionable uses of funds.

At the time of our review, the U.S. Supreme Court had agreed to review a court decision, which would require changes in how HEW regulated and administered emergency assistance. A decision was expected by June 30, 1978.

We recommended that the Secretary of HEW:

--Pursue efforts, through the Congress if necessary, to resolve the definitional and interpretational problems hindering the operation of the Emergency Assistance program.

--Develop uniform guidelines for administering the program based on an appropriate definition of emergency assistance and in line with the U.S. Supreme Court's expected decision.

--Monitor States' programs on a continuing basis to insure compliance, once definitive criteria for emergency assistance and uniform guidelines are developed.

We recommended that the Congress should consider whether the Emergency Assistance program should continue because:

--In fiscal year 1976, seven States accounted for 87 percent of the program expenditures and two of the seven accounted for 50 percent.

--As of September 1977, 30 States were not participating in the program., Nineteen of them provided emergency assistance under their own programs or the special needs category of the AFDC program, and this was one reason they did not participate. Of the 19 States, 9 said they did not participate primarily because they provide emergency assistance by these other means.

We further recommended that, if the Congress determined the program should continue, it should review the positions of HEW and the courts, including the U.S. Supreme Court, concerning eligibility and the type and extent of emergencies covered. It should then, if necessary, amend the legislation to clearly indicate congressional intent.

60

. ; ,1 , n : : . ,

‘, ..’ .‘, ,_ .~ : :

) . - . .

. . : : . ‘1 , .

8’ : ‘, , -

:’

.,,a> ,‘I

: \ ; . _. , . ..’

j,

, -‘,

: . I I ; .i , ; - ,


payments. HEW did not consider the administrative costs that would be associated with implementing corrective actions. In addition, States generally did not conduct cost effectiveness studies before starting corrective actions, although they were required to do so by HEW.

We recommended that, to improve the administration of the quality control program, the Secretary of HEW revise the Department's basis for determining accomplishments resulting from States' efforts to reduce errors. HEW revised its method for determining savings by using statistical tests of significance and has revised its methodology in accordance with our recommendation.

We recommended that HEW base its reporting of State errors on dollar amounts rather than on case error rates. Currently, HHS is reporting State errors with a focus on dollar amounts rather than case error rates, although it continues to develop both case and payment error rates.

We also recommended that HEW help States identify corrective actions that can be demonstrated to be cost effective. HEW had a study underway to examine the cost effectiveness of corrective actions at the time of our review. The results of this study showed that actual program savings resulting from reduced erroneous payments substantially exceed the administrative cost attributable to quality control and corrective action efforts. However, the study also showed that costs and difficulties were encountered in attempting to isolate the cause and effect relationship between multicorrective actions and specific error reductions so that calculating net savings on an ongoing basis would be impractical.

In addition to continuing efforts to provide technical assistance on ways to reduce,errors, such as issuing publica- tions, holding workshops, and engaging contractors, HEW devel- aped, in line with our recommendation, an expanded program focusing on key areas in the use of proven effective corrective actions, such as the use of "error-prone" profiles for cases likely to have errors, monthly recipient reporting systems, and consolidating assistance standards.

HRD-80-80

We reviewed the AFDC quality control system as a result of a September 11, 1978, request of the Chairman, Senate Committee on Finance.


HRD-76-164

HEW had attempted to encourage error reduction by requiring States to implement quality control programs and issued regulations that provided for financial sanctions or penalties against States with high error rates. HEW's quality control system was established in 1973 to identify and measure incorrect payments for the purpose of giving management information for developing corrective action to reduce errors. The system had been operated mainly by States, and the HEW regional quality control staffs had been reviewing and monitoring State systems. HEW headquarters staff oversaw the efforts and compiled national error rate statistics.

Sixteen States and one county challenged the AFDC sanc- tion regulation, contending that it penalized States for errors they could not reasonably be expected to correct and that the specific sanctions were arbitrary. The courts ruled in favor of the States that the specific sanctions imposed were unenforceable, but that under the Secretary's rulemaking power, HEW could impose reasonable sanctions supported by factual bases rather than an arbitrarily established tolerance level.

We recommended that the Congress determine the control that would best provide desirable financial incentives and should enact legislation to establish such incentives to effectively control AFDC payment errors. We stated that the Congress, in its development of such legislation, should seek HEW's assistance to determine an appropriate and feasible incentive. We believed that such legislation should provide for using a payment error rate as the basis for setting goals for measuring States' accomplishments in reducing errors.

The Congress took positive action by including a formula in the Social Security Amendments of 1977 by which States that reduced their quality control error rates below 4 percent could participate increasingly in the Federal share of the money saved. According to HEW officials, three States-- Nevada, North Dakota, and Oklahoma--had reduced their error rates sufficiently to be eligible to participate in the Fed- eral funds saved during 1978.

Since the quality control program was initiated in 1973, HEW had continually overstated the program's accomplishments. Savings estimates resulting from error reductions were not based on valid statistical projections and included actions which did not necessarily produce direct savings in welfare

62


Recommendation to the Appropriations Committees

We recommend that the House and Senate Appropriations Committees retract the conference committee directive for Federal fiscal sanctions again&t States based on the AFDC quality control error rates.

."*S, m"-ENT pBINTINO OFFICE : 1980 O-341-844/504


Our report recommended several HHS actions needed to make the quality control system more effective. HHS officials said that efforts underway would bring about improvements, but because of the recency of our report (July 18, 1980) we did not follow up to determine the status of HHS' actions.

Our report also pointed out that congressional action had created an adversary relationship between the Federal Government and States at a time when cooperative effort was needed to reduce errors. The congressional conference on the 1979 supplemental appropriations bill (Public Law 96-38) attempted to encourage States to reduce errors by directing HEW to withhold Federal funds for erroneous payments above certain tolerances based on quality control findings. The congressional conferees directed HEW to issue regulations requiring States to reduce the AFDC error rate to 4 percent by September 1982 (in equal amounts each year beginning in fiscal year 1980) or to lose Federal matching funds associated with erroneous payments in excess of the target. In

, January 1980, HEW issued final rules to implement the sanctions directive.

We believe that using the quality control system as the basis for sanctions limits the system's value as a means for improving payment processes. Because a high error rate will result in sanctions, there is an incentive to identify fewer errors. Instead of sanctioning States, we believe that the Federal Government should provide more assistance in error reduction efforts. Therefore, we recommended that the House and Senate Appropriations Committees retract the conference committee directive for Federal fiscal sanctions against States based on the AFDC quality control error rates.


We recommend that the Congress, in cooperation with HHS, determine the controls that would best provide additional appropriate and feasible financial incentives and enact legislation to establish them to effectively control AFDC payment errors. The Congress should consider HHS' study to determine an ultimate error rate goal in establishing the needed financial incentives.

64

. .

‘. i, , *

AN EQUAL OPPORTUNiTY EMPLOYER

UNITED STATES GENERAL ACCOUNTING OFFICE

WASHINGTON, D.C. 20548

OFFICIAL BUSIWESS

PENAL171 FOR PRIVATE us%c.u@

PoarAoc ANO PILLS P*lD _ U. ,. GRNRRAL ACCOUNTING OFClCf

THIRD CLASS